Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrationsand photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrightmaterial had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuingfrom left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

U-M-I University Microfilms International A Bell & Howell tntormancn Company 300 North Zeeb Road. Ann Arbor, M148106-1346 USA 313/761-4700 800/521-0600

Order Number 933492'1

Colonization of seagrass leaves: A model biological system for the study of recruitment in a marine environment

Michael-Taxis, Teena, Ph.D.

University of Hawaii, 1993

U·M·! 300 N. Zeeb Rd. Ann Arbor, MI 48106

COLONIZATION OF SEAGRASS LEAVES;

A MODEL BIOLOGICAL SYSTEM FOR THE STUDY OF

RECRUITMENT IN A MARINE ENVIRONMENT

A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAII IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

BOTANICAL SCIENCES (BOTANY)

AUGUST 1993

Teena Michael-Taxis

Dissertation Committee:

iii

------ACKNOWLEDGEMENTS

I deeply appreciate each committee member including the late Dr. Sanford Siegel for his or her role in the

development of this resea~ch. The insights and efforts of Dr. Celia smith in particular as well as Drs. Isabella Abbott, Kim Bridges, Robert Kinzie III and Harry Yamamoto were invaluable in preparation of the text and extending questions that furthers the research. Tina Corvalo and the Biological Electron Microscope Facility assisted in many aspects of the ultrastructural analysis. This research was supported by the Department of Botany, University of Hawaii, the U.S. Office of Naval Research Grant No. N00014-90-J-1932 and Hawaii Natural Energy Institute, University of Hawaii, Honolulu, Hawaii. I extend special appreciation to Rick Hanna for his multifaceted role in development of this dissertation. I gratefully acknowledge the support of Harriet Matsumoto, Dr. Gerry Carr, Gerry Ochicubo and Dora Tsuha from the Botany Department. I sincerely thank my friends and family, Jan Taketa, Sue Douglas, Phillip Moravchic, In Sun Kim, Chris Omeara, Jane Dawson, Marie Bruegman, Naomi Phillips, Luis Vega and my mother and daughter, Milledge and Teale for their patience, help and challenges.

iv ABSTRACT

Leaves of the Hawaiian seagrass, Halophila hawaiiana Doty and stone are a base for diverse epiphyte communities. The leaves were considered a model system for the study of patterns in colonization. The anatomy of the leaves, as substrate for colonization, was documented prior to investigating patterns and processes of colonization along a gradient of wave exposure. Ultrastructural assessment of mature leaves revealed details of epidermal, ground and vascular tissues that extends our knowledge of the genus. Cell wall ingrowths with invaginated plasmalemma characterized the epidermal cells and were most elaborately developed in the upper leaf surfaces. Chloroplasts, mitochondria, endoplasmic reticulum and dictyosomes are commonly associated with the ingrowth regions. structures indicative of symplastic and apoplastic systems were detected. Mature, twelve day old leaves showed ultrastructural modifications of when colonized by specific epiphytes. These modifications included: 1) distinctive elaborations of the cell wall ingrowths and abundant secretory organelles when colonized by crustose coralline algae, 2) disruption of the fibrillar cell wall, osmiophilic droplets, vesiculate membrane-bound structures and altered ingrowth regions as well as reduced numbers of chloroplasts and mitochondria when colonized by specific bacteria. The

v distribution of these epiphytes followed a wave exposure gradient. Colonization by filamentous red algae, cyanobacteria and bacteria in microcolonies occurred in both sites and did not alter the leaf ultrastructure. Distinctive polysaccharides were observed at the microbial cell surfaces; these molecules may provide adhesion between host and epiphytes. A possible mechanism(s) driving patterns in colonization and recruitment was examined via novel use of specific lectins as probes for newly emergent seagrass leaves and an artificial substrate. Distributions and identities of several naturally occurring glycoconjugates were resolved in films that formed between one and three days on glass slides. In contrast, no film nor glycoconjugate sites were visible on surfaces of young seagrass leaves. The spatial and chemical heterogeneity of settlement cues on substrates may generate microscale patterns in "lock and key mechanism(s)". Surface glycoconjugates appear to provide a cue for recognition by settling stages that may provide an initial step that influences ultimate community features.

VI TABLE OF CONTENTS

Acknowledgements ••...... ••••••••••••••••••••••••••••••.••. i v Abstract v

List of Tables xi

List of Figures xii List of Abbreviations xviii Chapter 1: Literature Review••••••••.••••••.••••••••••.•.•.1

Background 1

Introduction to Plant community Ecology•.•..•..•.••.••• 2

Hawaiian Seagrass Biology, Taxonomy and Ecology•.•...•. 6

Seagrass Leaves as Host Substrates•••.••••••••••••••••• 9

Epiphyte Communities••••••••••••.••••••••••••••••••••• 13

Epiphyte Influences on Hosts••••••••••••.••.•.••.•.•.• 17

Establishment of Seagrass and Epiphyte Associations•.• 19

Forces of Attachment..•....•...... •...... 25

Surface Properties and Modifying Films••.••.••••••..•• 28

Surface Texture 29 Surface Tension 30

Hydrophobic Surface Properties.•••.•...•..•.••.•.•.• 31

Surface Charge 32

Surface Modifications•.•••.•••••••••••.•.••....•.... 33

Glycoprotein-mediated Cell Surface Interactions•.... 34

Focus on Halophila hawaiiana as Host for Epiphytes...• 37

Literature Cited 38 Chapter 2: Leaf Ultrastructure of the Hawaiian Seagrass, Halophila hawaiiana Doty and Stone.... 56

Abstract 56 vii Introduction 57

systematics 57

Anatomy •.••••••••••••••••••••••.•••••.••••.•••••••.••• 60

Materials and Methods. • ••••••••••••••••••• 63

Collection of Halophila hawaiiana...... 63 Sample Preparation for Transmission Electron Microscopy . . 64 Observations by Light Microscopy...... •.••.•...... •... 65

Results 66

The Epidermis 67

Ground Tissue 69

Vascular Tissue 70

Discussion 73

Conclusions. . 82

Literature cited. •••.• 102

Chapter 3: Ultrastructure of Seagrass and Epiphyte Interfaces from Wave Exposed and Sheltered Subtidal Habitats••••••..•••...••..•.••••...... 107

Abstract 107

Introduction. .108

Epiphytes... • 109

The Host and Epiphyte Interface••...... ••...•....•... 111 Environmental Influences on Settlement of Epiphytes 114

Materials and Methods....•..•...... •... 116

Sampling and site...... 116

Sample Preparation for Electron Microscopy 116

viii Epiphyte Population Counts..••••••••.•.••••••••••.••. 118

Results••• .120

Part A. Technical and Qualitative Evaluation of Epiphytes and the Host••••••••••••••••••••.•••••• .120

Fixation Results••••• .120

A Contrast of Seagrass and Epiphyte Populations/Site••••••••••••••••.••• • ••• 122

Part B. Qualitative Observations on Epiphytized Seagrass Leaves ••.•..•••••.•.•••.•••...... 125

The Epiphytes. .125

The Host and Epiphyte Interface. .127 Discussion•• ...... 132 Conclusions. .145

Literature cited. .173

Chapter 4: Lectins Probe Molecular Films in Biofouling: Characterization of Early Films on Living and Non-living Surfaces•••.••••••• .182 Abstract••••• ...... 182 Introduction...... 183 Materials and Methods. .186

Laboratory Films•.•• .186

Natural Films on Glass. .187

Natural Films on a Living Surface. .187

Application of Lectins••••.••••••• .187

Microscopy. .189

Results•••••• .189

single Lectin Studies in Artificial Films. .189

ix Lectin Studies of One and Three-day Natural Films on Glass Slides . • •• 190 Lectin Studies of a Newly Emergent Living Substrate 192

Discussion•• ...... • •••• 193

Conclusions. • •••• 197

Literature cited. • .203

Chapter 5: Synthesis. •• 208

x LIST OF TABLES

Table Page

3.1. Preservation of Eukaryotic and Prokaryotic Cells and Cell Products•.••••.••••••••••••••••. 147

3.2. A Profile of Halophila hawaiiana and its Epiphytes; Number of Epiphyte Cells/Host cell. N = 100 host cells 148

3.3. A Profile of Halophila hawaiiana Surface Cell Components; Number of Chloroplasts and Layers of Cell Wall and Plasmalemma Elaborations••••.. 149

3.4. Assessment of Significant Differences within Categories (95% confidence interval) ••.••••.••. 150

4.1. Excitation and Emission Wavelengths for Fluorescent Chromophores Available with Lectins - 199

4.2. Sugar Specificity and Molecular Characterizations of Common Lectins••••••••••.•• 200

xi LIST OF FIGURES

Figure

2.1. Habit of Halophila hawaiiana•.•.••.••..••••••••••• 85

2.2. Leaf Morphology and venation Pattern•••••••••••••• 85

2.3. Leaf of Halophila hawaiiana, Transverse section 87

2.4. Surface View of Leaf Cell Pattern••••••••••••••••• 87

The Epidermis

2.10. Symplastic Connections Within the Epidermis I .....91

2.11. Symplastic Connections within the Epidermis II....91

2.12. Symplastic Connections within the Epidermis III...91

2.13. Transverse Vein Ground Tissue I ...... 92

2.14. Transverse Vein Ground Tissue II...... 92

2.15. Transverse Vein Ground Tissue III...... 92

2.16. Transverse Vein Vascular Tissue I ...... 94

2.17. Transverse Vein Vascular Tissue II...... 94

2.18. Transverse Vein Vascular Tissue III...... 94

2.19. Transverse Vein Vascular Tissue IV...... 94

2.20. Transverse Vein Vascular Tissue V •••••••••••••••• • 94

2.21. Transverse Vein Vascular Tissue VI ...... ••.... 96

2.22. Transverse Vein Vascular Tissue VII ...... 96

2.23. Marginal Vein Intercellular Connections..••.••.••• 97

xii 2.24. Central Vein Vascular Tissue I ••....••••••..•••••• 97

2.25. Central Vein Vascular Tissue II••.•..•••••.••.•••.97

2.26. Central Vein Apoplasm I •••••••••••••••••••.••••••• 99

2.27. Central Vein Apoplasm II•••••••••.•..•••••.••.••••99

2.28. Central Vein Apoplasm III••••.••.••••••••••••.•••. 99

2.29. Central Vein Apoplasm IV...... •..99

2.30. Chloroplast Morphology in the Epidermal Cells.••• 101

2.31. Chloroplast Morphology in Ground Tissue I •.•••••• 101

2.32. Chloroplast Morphology in Ground Tissue II•.•.••• 101

2.33. Chloroplast Morphology in Ground Tissue III••.••• 101 3.1. Multilayered Epiphyte Community on Adaxial Seagrass Surface, Wave Exposed site I •••.•••••• 152

3.2. Multilayered Epiphyte Community on Adaxial Seagrass Surface, Wave Exposed site II•••••.••• 152

3.3. Multilayered Epiphyte community on Adaxial Seagrass Surface, Wave Exposed site III•••••••• 152

3.4. MUltilayered Epiphyte Community on Adaxial Seagrass Surface, Wave Exposed site IV••••••••• 152

3.5. Multilayered epiphyte community on adaxial seagrass surface, wave exposed site V•••••••••• 154

3.6. Pennate diatoms adherent to seagrass surface, wave exposed site 154

3.7. A diatom frustrule remains attached to the seagrass surface by an apparent attachment plug following death of the cell••.•••••••••••• 154

3.8. Unepiphytized adaxial seagrass surface, a control for assessing cellular modification coincident with epiphytes.....•.••••..•••.•..•• 154

3.9. Epidermal cell modifications coincident with coralline red algal encrustation, wave exposed site I 156

xiii 3.10. Epidermal cell modifications coincident with coralline red algal encrustation, wave exposed site II 156

3.11. Epidermal cell modifications coincident with coralline red algal encrustation, wave exposed site III 156

3.12. Epidermal cell modifications coincident with bacteria within cell wall, wave exposed site•.•156

3.13. Higher magnification of bacterium within seagrass cell wall, wave exposed site I .••••.••158

3.14. Higher magnification of bacterium within seagrass cell wall, wave exposed site II•.•••••158 3.15. Overview of bacteria on leaf, wave exposed site 160

3.16. Bacteria attached to seagrass leaves, variation in glycocalyces, wave exposed site I 160 3.17. Bacteria attached to seagrass leaves, variation in glycocalyces, wave exposed site II 160

3.18. Bacteria attached to seagrass leaves, variation in glycocalyces, wave exposed site III 160 3.19. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site I 162

3.20. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site II 162

3.21. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site III 162

3.22. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site IV..•...... 162

3.23. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed

site V III •• 0 0 0 0 III III III • 0 •• 0 III • 0 0 III III 0 III III 0 0 III III III III 0 III 0 III 162

xiv 3.24. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site VI .•...... ••.... 162 3.25. Epiphyte community on adaxial seagrass surface, wave sheltered site I •••.•••..•••••.••.••••.... 164 3.26. Epiphyte community on adaxial seagrass surface, wave sheltered site II•••.••.••.••••.•••••••••. 164 3.27. Epiphyte community on adaxial seagrass surface, wave sheltered site III•.•••.••.•••.••••••••.•. 164 3.28. Epiphyte community on adaxial seagrass surface, wave sheltered site IV •••.••..••••••••••••••..• 164 3.29. Epiphyte community on adaxial seagrass surface, wave sheltered site V••••••...••••••••••••••.•• 164 3.30. Diatom adherent to the seagrass, carbohydrate containing material adjoins diatom to surface 166 3.31. Profile of epiphyte and seagrass interface••••••. 166 3.32. Detail of extracellular material and a bacterium between diatom and seagrass surface•• 166 3.33. Extracellular material between a diatom and bacterium, and the seagrass surface••••.••••••. 166 3.34. Overview of bacteria on the seagrass, wave sheltered site 168 3.35. Hemispherical bacterium adnate to the seagrass surface 168 3.36. Coccoid bacterium adherent to the seagrass by fibrous glycocalyx 168 3.37. Bacteria with heterogeneous glycocalyx composition attached to or associated with degrading seagrass wall, wave sheltered site...168 3.38. Bacterial infection at lysed seagrass surface, wave sheltered site•..•••.•.....•••..••..•••... 168 3.39. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site I •...•.•••...... ••.•.•••.•• 170

xv 3.40. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site II.•••..•••••..••••••.••••• 170

3.41. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site III••••.••••••.•••••••••••• 170

3.42. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site IV.•••••..••••••••••••••••• 170

3.43. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site V•.•.••••••....•••••••••••• 170

3.44. Diverse bacteria attached to the seagrass with tannic acid positive glycocalyx components, wave sheltered site VI .•.•••••••••••••••••••••• 170

3.45. Diverse bacteria associated with the seagrass, wave sheltered site I •.•••••••••••.•••••••••••• 172

3.46. Diverse bacteria associated with the seagrass, wave sheltered site II.•••••.•••••••.•••••••••• 172

3.47. Diverse bacteria associated with the seagrass, wave sheltered site III•••.•..••••••••••••••••• 172

3.48. Diverse bacteria associated with the seagrass, wave sheltered site IV•••.•••.••••••••••••••••• 172

3.49. Diverse bacteria associated with the seagrass, wave sheltered site V•.••.•.••..••••.•••••••••• 172

3.50. Diverse bacteria associated with the seagrass, wave sheltered site VI•..•••..••••••••••••••••• 172

4.1. Positive Signal from Con A + FITC Bound to a Glucose Film 201

4.2. Autofluorescence from Unstained Three-Day Films on Glass 201

4.3. Microorganisms in a Three-Day Film on Glass Localized by DAPI Staining.••••••..••••••.••••• 201

4.4. Distinct spatial Pattern of Glycoconjugates Distribution in the Matrix of a One-Day Film on Glass 201

xvi 4.5. Circular Pattern of Distribution of a One-day Film on Glass••••.•.•••...••..••••••.•• 201

4.6. Complex Spatial Pattern in a Three-Day Film on Glass...... •...... •...... 201

4.7. Magnified Area of Fig. 4.6••••••••••••••••••••••• 201

4.8. Extracellular Polymeric Secretions of Bacteria in Three-Day Films on Glass••••••••••• 201

4.9. Autofluorescence by Cells of Newly Emergent Leaves of Halophila hawaiiana.••••••.••••••••••201

4.10. DAPI stained Nuclei of Seagrass Cells•••••••••••• 201 4.11. Epidermal Cells as Revealed via Limulin (+ FITC} 201

4.12. Epidermal Cell Walls as Revealed via Helix Lectin + FITC 201

xvii LIST OF ABBREVIATIONS

Ae air canal sp electron dense spheres As air space st sieve tube b bacteria T thylakoids Bs bundle sheath t tunneling bacteria ea capsule, glycocalyx Tv transverse vein eb cyanobacteria u upright ee crustose coralline red V vacuole algae ve vessicle Ce companion cell vo void around bacterial el chloroplast cell ep clay silt particles Vp vascular parenchyma cv central vein w bacterial wall d diatom X xylem e envelope (plastid) Ep epidermal cell er endoplasmic reticulum f attachment fringe fL foliage leaf q cell ghosts ql glycogen-like granules b PHB-like inclusions I ingrowths (cell wall) In internode iw inner wall I lipid m mitochondria mb microbody me microcolony mu mucilage Mv marginal vein N nucleus n nucleoplasm no node o inclusions om organic matter ow outer wall P plasmodesmata Pa parenchyma cell pe petiole pI diatom attachment plug pm plasmalemma Pp phloem parenchyma pr sieve tube plastid r ribosomes ra red algae xh rhizome ro root S starch grain sL scale leaf xviii CHAPTER 1

LITERATURE REVIEW

Background A community can be broadly defined as an assemblage of species that occurs together in space and time. A primary focus of community ecology is understanding the spatial distribution of these assemblages and the interactions among species and abiotic features of the environment. While the ubiquity of species interactions is obvious, it is less clear if interactions within a community impart an organizational structure to the community. This leads to a central question: how do settlement processes in early stages of community development contribute to later community structure? This review discusses pertinent background literature in nine major sections. with this discussion, I am organizing the background for evaluating a long-standing but still untested paradigm in community ecology: the early processes of community development direct subsequent species composition and dynamics. For a propagule, these early processes include site availability, selection, adhesion, and attachment. My evaluation of the paradigm focuses on site selection and settlement on seagrass as a model biological surface.

1 Introduction to plant community ecology Many researchers have worked, using a variety of approaches, to define the concept of a community. Some plant ecologists who focus on terrestrial ecosystems, consider the resident species and their interactions to be integral to the definition of a community. Proponents of this view consider the plant community to be analogous to an organism. For example, Clements (1916, 1928) proposed that a plant community may be viewed as an organism with discrete units and boundaries that develops in a structured successional pattern. The Clementsian view is that a community matures through repeatable serial stages to a predictable and stable endpoint, the climax community. The community functions as a whole and is greater than the sum of the parts. Tansley (1920) also regarded the plant community as a whole but emphasized the inclusion of both species that are "dependent" and found only within the entity, and independent species that can establish in other communities. In contrast, Gleason (1926, 1939) emphasized the individuality of each plant community as a random assemblage of organisms that occurs in similar environments; because the community depends on surrounding vegetation and the environment, constant spatial and temporal changes in those aspects lead to each community being unique. Alternatively, other researchers maintain that association does not imply " .••harmonious concurrence of

2 diverse activities working towards a common endll but the co existence of forms that live side by side for their " ••. own exclusive profit, II (Flahault and Scroter 1910). As Mueller Dombois and Ellenberg (1974) point out, emphasis on the individuality of a plant does not exclude relationships among the plants of a community. Species can compete for the same resource and non-competing species may differ in requirements or seasonality. Species can depend on a particular niche created by a dominant species or a species can require a specific host (Walter 1964, 1971). Theories of marine benthic community organization stress the importance of competition and predation, environmental gradients, disturbance, and early settlement events. Settlement rates and patterns have been shown, in some locations, to depend on propagule supply and substrate suitability (Connell 1985). variation in the nature of the propagule and the rate of recruitment can contribute to species diversity, community structure and dynamics (Keough and Downes 1982). Roughgarden et ale (1986) developed a theory of IIsupply side ecology" that stresses the relationship between potential recruit abundance, resource limitation and competition in some communities. Taxonomic composition of the gene pool, hydrodynamics, physical characteristics of the substrate, and disturbance interact in recruitment and colonization (Palmer 1988). The success of settlement, germination and recruitment processes by

3 propagules determines algal establishment and may influence species distribution (eg. McDermid 1988). Active and passive mechanisms are involved in recruitment from the water column and colonization of benthic surfaces. The velocity boundary layer, defined as the layer of water near where the current velocity has been reduced by the surface (Lobban et ale 1985), can influence the surface microenvironment for settlement. If a propagule is retained in the boundary layer of a substrate long enough to settle, adhesion may occur. In later sections, I discuss the forces of attachment and properties of surfaces as background to interfacial interactions that may direct settlement on surfaces in the marine environment. The leaves of seagrasses offer epiphytes large surface areas for these interactions and consequently a novel opportunity for research on the processes of epiphyte community development and structure. Living substrates, like seagrass leaves, potentially offer different constraints for colonists than non-living substrata. As such, a living surface might be expected to be biologically dynamic. For example, physiological fluctuations may occur on a leaf that create a mosaic of microhabitats for colonists. Surfaces of a biological host may change in physical and chemical characteristics with growth, onset of reproduction, and senescence. Each host

4 may possess specific surface chemistries that influence the type and distribution of colonists. Conversely, colonists can influence the host. For example, shading by epiphytes can limit the depth distribution with depth in some macrophytes. Epiphytes can affect the fitness of their hosts by 1) reducing host growth and fecundity, 2) increasing host vulnerability to dislodgement and loss from excessive hydrodynamic drag, 3) affecting host vulnerability to grazing and 4) changing the potential for metabolic processes that occur at the host surface by altering flow patterns at the boundary layer. The potential for interaction between a host and colonist is great because of their physical contact. Interactions between hosts and epiphytes may encompass many levels of ecosystem function. In this review, I discuss some of these levels by examining patterns in species associations, substrate selection, adhesion and settlement as these topics relate to epiphytes on seagrasses. This system has been considered a natural model for the study of these processes between sessile marine algae and invertebrates in other tropical systems (Borowitzka and Lethbridge 1989). I focus here on the Hawaiian seagrass, Halophila hawaiiana Doty and stone as a surface for colonization. In the following sections I describe patterns in epiphyte community development and probe possible bases for these patterns.

5 Hawaiian seaqrass biology, taxonomy and ecology The genus Halophila Thours in the family Hydrocharitaceae is widely distributed in all tropical seas and extends into SUbtropical and warm-temperate waters (den Hartog 1970). The growth of all Halophila species results in rapid horizontal spreading over the bottom but not accumulation of biomass (den Hartog 1970). The leaf-area index of Halophila is low and little biomass accumulates in perennial rhizomes. Colonization is rapid when light conditions are not limiting (Josselyn et ale 1986). Plants root easily from fragments and can tolerate being covered by sand or mud. Halophila species seem well adapted to unstable substrata (den Hartog 1970, Birch and Birch 1984). Apparent adaptations to physical disturbance and grazing include rapid growth rate and leaf turnover (Herbert 1984, Unabia 1984). The general description of Halophila which includes small stature and sparse growth, are not features usually associated with sediment stabilization. Yet, in areas of increasing sedimentation, the growth pattern enables plants to keep pace with rapid coastal rates of deposition (Doty and stone 1966). In calm conditions the blade density is high: sediment accumulation leads to ridges or mounds in the sea bottom (den Hartog 1970, Unabia 1984). Populations of Halophila hawaiiana and the closely related H. ovalis (R. Br.) Hook F. from Samoa and Tonga, have great morphological variability in plant size, leaf-

6 shape, nervation, length of the petiole, length of transition of the leaf to a petiole, leaf texture, rhizome thickness and numbers of roots per node. Although obvious phenotype variation is observed, morphologically distinct forms co-occur without nearby transitional forms, suggesting a genotypic basis for these variations (Tomlinson 1980). Halophila hawaiiana was separated from the R. ovalis "complex" by Doty and stone (1966) based on spacing and degree of branching from the midvein of the lateral veins and smaller leaves, the latter being one of the characteristics of H. ovalis complex. Another species distinction is the number of ovules (12-15), fewer than in R. ovalis (Sachet and Fosberg 1973). McMillan and Williams (1980) substantiated the genetic basis for separating R. hawaiiana by demonstrating the occurrence of unique isozymes, based on a study of seven enzyme systems. A fourth feature is the absence of flavone sulphates that occur in the R. ovalis complex and many other seagrasses (McMillan et ale 1980). Hawaiian seagrass meadows are comprised of a single seagrass species, H. hawaiiana that is endemic to the Hawaiian Islands. The R. ovalis complex, from which the Hawaiian species was removed, is "eurybiotic"; it extends from mid-tidal level to at least twelve meters deep and occurs on all kinds of substrata from coarse coral rubble to soft mud. Populations of H. hawaiiana are found from nearly

------_._. ------intertidal to a depth of ten meters. Their distribution includes sheltered to semi-exposed areas on sand or mud. This species generally forms patches although occasionally large continuous meadows are found. The patches are dynamic; they grow in size, move with the substrate, and disappear. In some protected areas, however, patch size and location remain constant over several years (Herbert 1984, Unabia 1984). Specific salinity and temperature requirements for the Hawaiian seagrass are not known. Carbonate sediments are often associated with tropical seagrasses such as Halophila. These sediments contain: 1) low amounts of organic matter; 2) high redox conditions; and 3) have low concentrations of fermentation products produced by anaerobic bacterial metabolism. Wetzel and Penhale (1979) have shown that a low rate of dissolved organic carbon (DOC) excretion correlates positively with high photosynthetic efficiency. High rates of carbon uptake, photosynthesis, quantum yield and transport to leaves characterize pioneer seagrasses in general. Photosynthetic characterizations of H. hawaiiana are not available. The grazer-relations of H. hawaiiana include extensive grazing by a monophagous mollusk, Smaragdia bryanae (Unabia 1984). Smaragdia bryanae can occur in densities greater than 500 per square meter in H. hawaiiana meadows. Consumption by this herbivore is estimated to be

8 approximately 11% of the standing crop or 45% of new leaf production as measured at 10 sites on O'ahu. Leaf turnover in H. hawaiiana has been estimated to be 14.7 days (Herbert 1984). This estimate is based on the relationship between the plastochrone interval, or the distance between successive nodes, and leaf area. This turnover rate is similar to the rate of another pioneer species and significantly greater than rates of two climax seagrass species (68-270 days) (Herbert 1984). The studies of Doty and stone (1966), Herbert (1984) and Unabia (1984) provide insights into taxonomic distinctions, timing of leaf production and flowering, and impact of a grazer association. However, many major questions concerning the autecology and community ecology of this species remain unanswered. Questions from those disciplines and within the focus of this study include: 1) what factors distinguish this seagrass as a host? and 2) what can we learn of colonization in the marine environment from this seagrass?

Seagrass leaves as host substrates Seagrass leaves provide unusually uniform surfaces for epiphytic growth in marine environments (Feldman 1937, Penhale 1977, Kenworthy et ale 1989, Duarte 1989). The leaves of a seagrass meadow reduce current speeds and propagules are retained in seagrass meadows. This baffle effect is thought to enhance recruitment rates by increasing

9 the window of opportunity for settlement (Hootsmans and Vermaat 1983, Peterson 1986). Thus, settlement is generally greater in seagrass meadows compared with surrounding areas (Hootsmans and Vermaat 1985, Wilson 1990). While the seagrass population influences water flow in the meadow, the form of the leaves influences hydrodynamic forces at the boundary layer (Cinelli et ale 1984, Lobban et ale 1984). From this, leaf morphology may be a critical factor indirectly controlling colonization, diversification and biomass of epiphytes. Seagrass leaf morphologies (den Hartog 1970) have been used to categorize seagrasses as substrates (Borowitzka and Lethbridge 1989). As substrata for colonization and SUbsequent community development, features of seagrass leaf structure and function determine the nature of the microhabitat. The leaf surface consists of a cuticle and cell walls that enclose photosynthetic, metabolically active epidermal cells. Two basic cuticle morphologies have been reported in the literature. The first, reported by Gessner (1968) is thin, porous and highly structured cuticle. The second type is thin and amorphous (Barnabas et ale 1977, Doohan and Newcomb 1976, Jagels 1973 1983, Kuo 1978). Subcuticular cavities have been observed in one seagrass and may function in ion absorption, secretion and/or storage (Barnabas et ale 1977, 1982). The cell walls are composed mainly of cellulose, hemicellulose and pectin (Kuo et ale 1990). The

10 cell wall composition of Halophila includes uronic acid as well as cellulose (Waldron et al. 1990) which may function in ion-exchange as do many acidic polysaccharides in the seaweeds (Chapman 1979, Waldron et al. 1990). The cellulose fraction increases with age, and non-cellulosic sugars make up 8-15% of the cell wall carbohydrates. In general, little information is available concerning the cell wall biochemistry although it is the essential interface between these cells, molecular interactions, epiphytes and the marine environment. The seagrass leaf epidermis is the site of gas and nutrient exchange as well as photosynthesis and osmoregulation. These cells have "transfer cell" configuration from an invagination of the plasmalemma and may function in cell secretion, and absorption (Gunning and Pate 1969, Gunning 1977) and osmoregulation (Jagels 1973, 1983). The surface chemistry and ultrastructural detail of the Hawaiian seagrass have not been studied. The fundamental anatomy of host and epiphyte interfaces are yet undescribed for most host and epiphyte complexes, including that of Halophila hawaiiana. Phenolic compounds are prevalent in marine seagrasses (Cariello and zanetti 1979a, Cariello et al. 1979, Zapata and McMillan 1979). Functionally, these compounds can inhibit epiphyte growth (Sieburth and Conover 1965, Ryland 1974, Harrison and Chan 1980, Harrison 1982). Chicoric

11 acid, the dominant phenolic compound in the young leaves of Posidonia (cariello and zanetti 1979b), retards epiphytic colonization (Mazzella et al. 1981). Tannin cells are absent in Halophila yet compounds with characteristics of condensed tannins occur in this genus (McMillan 1984). Sulfated phenolic compounds in Halophila play a role in osmotic adjustment to the marine environment (McMillan et ale 1980), but their role remains untested for the Hawaiian species. Exuded compounds have been shown to affect epiphyte community composition and dynamics in several ways. Antibiotic compounds are released by some seagrasses (Burkholder 1973) and regulate the diversity and composition of the microbial epiphyte community (Sand-Jensen 1977, Ducker and Knox 1984). Seagrass exudates can alter growth, either inhibiting or stimulating bacteria, diatoms and algae (Fitzgerald 1969, Harrison and Chan 1980, Harrison 1982, Novak 1984). Inhibitory exudates are most effective during the vulnerable attachment phase of colonization (Harlin 1980). We have no information on the exuded chemistry of H. hawaiiana. Seagrasses, in general, release up to 2% of photosynthetically fixed carbon as dissolved organic molecules (Byrlinski 1977, Penhale and Smith 1977, Wetzel and Penhale 1979, Moriarty et al. 1985). Dissolved organic matter (DOM) secreted by aquatic plants increases bacterial

12 densities in the phylloplane, or leaf areas (Hough and Wetzel 1975). Laboratory studies with labeled phosphorus have shown that leaves of Zostera marina L. leaves excreted 1 to 3% of the phosphorus absorbed by the roots (Penhale and Thayer 1980). Phylloplane epiphytes and rhizosphere, or root zone, bacteria capture at least some portion of released nutrients (Harlin 1973, Penhale and Smith 1977, Penhale and Thayer 1980, Moriarty and Iverson 1986). Harlin (1980) speculated that the effects of rhizosphere nitrogen metabolism may be observed in the phyllosphere via release of nitrogenous compounds exuded by a seagrass.

Epiphyte communities Seagrasses provide a physical substratum for bacteria, algae and invertebrates, access to irradiance for photosynthetic algae, and a source of organic carbon for bacteria, heterotrophic algae and animals (Harlin 1980). Hazards of biological substrata in general, for long-lived epiphytes, include seagrass sloughing at the distal end and as surface layers (Filion-Myklebust and Norton 1981, Moss 1982, Duarte 1989) as well as sand scour and whiplash of the leaves (Brawley and Xiugeng 1988). Epiphytes are ubiquitous in seagrass ecosystems (Borowitzka and Lethbridge 1989). They contribute significantly to the total energy budget and species diversity of near-shore ecosystems (Murray and Wetzel 1987).

13 In general, epiphytes are major primary producers in seagrass communities and important food sources for animals that reside in these seagrass beds. The structure of epiphyte communities has been viewed as a complex series of processes and interactions (Hudon and Bourget 1981, DIAntonio 1985). Epiphyte assemblages include autotrophic organisms (diatoms, cyanobacteria, macroalgae, encrusting algae), and heterotrophic organisms (bacteria, fungi, invertebrates). Each of these epiphytic organisms may be further epiphytized (Novak 1984), generating an overstory and understory (Hudon and Bourget 1983). Fountaine and Nigh (1983) operationally define the epiphyte community as epiphytic algae, any associated heterotrophic organisms, and detrital material. The ecological significance of various microalgal epiphytes is poorly documented (Murray and Wetzel 1987), yet the potential for interaction between a host and epiphyte as well as between epiphytes, is great because of their spatial proximity (Wetzel 1983). General ecological processes that describe the interactions we might expect include competition, predation, parasitism, mutualism, proto cooperation, commensalism and ammensalism (Pianka 1983). All of these concepts may be applicable to understanding host-epiphyte and epiphyte-epiphyte interactions. Epiphytes can increase the productivity in seagrass ecosystems (Jones 1968, Wood et ala 1969, Marshall et ala

14 1971, Thayer et ale 1975, Penhale 1977, Penhale and smith 1977, Borum and Wium-Andersen 1980, Zieman and Wetzel 1980, Morgan and Kitting 1984, Mazzella and Alberte 1986). This has been shown by estimates of photosynthesis and in situ

14C fixation (Jones 1968, Penhale 1977). Rates of photosynthesis by epiphytic algae can nearly equal that of its host on a gram/gram basis (Penhale 1977). Annual epiphyte production may be 10-40% of seagrass leaf production in a temperate ecosystem (Penhale 1977, Borum and Wium-Andersen 1980, Borum et ale 1984); epiphyte production in the tropics may be higher (Klumpp et ale 1989) although we have no estimate of photosynthetic performance for any component of Hawaiian seagrass meadows. In general, seagrass production may stimulate epiphyte production. Nutrient transfer from seagrasses to epiphytes has been demonstrated in many systems (McRoy and Barsdate 1970, McRoy et ale 1972, Harlin 1973, McRoy 1974, McRoy and Goering 1974, Harlin 1975, Brylinsky 1977, Penhale 1977, Penhale and smith 1977, smith and Penhale 1980, Libes and Boudouresque 1978). Dissolved organic matter released from the seagrass leaves into the surrounding water is reduced in the presence of epiphytic bacteria and algae (McRoy and Goering 1974). Transfer of carbon, nitrogen and phosphorus at the leaf surface is likely to enhance primary production by epiphytes (Libes and Boudouresque 1987). These

15 interactions remain uncharacterized for R. hawaiiana and its epiphytes. Patterns in epiphyte and host production have been assessed for some seagrasses. Epiphyte production (as biomass) on leaves of Zostera marina shows distinct patterns in each system studied. Trends include: 1) temperature influenced epiphyte biomass more strongly than did irradiance, nutrients or salinity (Brown 1962), 2) autotrophic and heterotrophic epiphyte biomass correspond to the seasonal nutrient regimes of orthophosphate and nitrogen in the water column (Borum 1985), 3) epiphyte biomass lagged behind host production (Mazzella 1983), and 4) epiphytic microalgal oxygen production increased as k. marina production declined yielding a sustained level of community production (Murray and Wetzel 1987). In another seagrass system, patterns in production by hosts and epiphytes showed positive correlations between biomass and temperature in both the host, Ruppia maritima L., and its epiphytes (Murray and Wetzel 1987). Others have reported an inverse relationship between production by a host Posidonia oceanica (L.) Delile and epiphyte biomass (Mazzella and Alberte 1986). Colonization on marine plant surfaces generally increases primary production for that ecosystem. The limitations and abundances of the environment combine with the metabolic optima of the particular species involved to determine complex outcomes

16 that are not fully predictable based on abiotic factors. The results from these and other studies in other systems provide a framework for evaluating potential synchrony and directionality in the interactions of the Hawaiian seagrass and its colonists.

Epiphyte influences on hosts Ducker and Knox (1984) have divided algal epiphytes into two classes based on degree of surface penetration: those that attach to the outer layers of the host (termed "holo-epiphytes") and those that anchor deeply in the host tissue (termed "amphi-epiphytes). Complex cytoplasmic connections and altered host metabolism have been described between "parasitic" algae and algal hosts (Goff 1982a, Goff and Coleman 1984). Defining, rather " .•.redefining our concepts of individual and sYmbiosis" (Goff 1982b) becomes increasingly necessary as information on surface-level interactions increases. At an ecological level, epiphytes may interact with hosts via 1) shading (eg. Sand-Jensen 1977), 2) altering the flow regime and reducing the diffusion rate of C02 and nutrients (Brawley and Adey 1981, Brawley and Xiugeng 1987) and 3) increasing the drag force on the blades (eg. Taylor and Lewis 1979, Borowitzka and Lethbridge 1989). Interpretation of these influences are problematic. For example, understanding the influence of epiphytes on light

17 harvesting by a host is complicated by both the problems associated with sampling methodologies and the complexity of the epiphyte-host interaction. In laboratory conditions, unepiphytized seagrasses generally tolerate a broad range of light intensities (McMillan 1980). Increased numbers of epiphytes can reduce seagrass photosynthesis by shading effects (Sand-Jensen 1977, Borum and Wium-Anderson 1980, Howard 1982, 1986, Bulthuis and Woelkerling 1983, Orth and Van Montfrans 1984, Howard and Short 1986, Josselyn et ale 1986). In the extreme, shading can contribute to the thinning or even destruction of meadows by both direct and indirect mechanisms (cambridge et ale 1986, Howard and Short 1986, Josselyn et ale 1986). Fouled leaves can grow more slowly as a result of direct mechanical damage (Howard and Short 1986). There is also evidence that epiphytes do not seriously impact their seagrass hosts. Mazzella and Alberte (1986) have shown that neither microheterotrophic nor autotrophic epiphytes significantly alter Zostera leaf photosynthesis (see also Kirchman et ale 1984). These authors note that the predominant light absorption by epiphytes is not in wavelengths absorbed by the light harvesting pigments of the host, and thus host and epiphyte are not directly competing for spectral qualities of light. Additionally, epiphytes can contribute nitrogen to hosts. Cyanobacterial associations with seagrasses can greatly increase ecosystem

18 production in nutrient-poor environments (Capone and Taylor 1980) via nitrogen fixation on surrogate seagrass leaves (Goering and Parker 1972). Although this finding has been disputed (McRoy et al. 1973, Patriquin and Knowles 1972), heterocystic cyanobacteria have been found on leaves as well as in the rhizosphere of seagrasses (Capone and Taylor 1980, Capone 1983) • Epiphytic colonization of leaves may act as an important evolutionary pressure, driving growth strategies of seagrasses to increase both the rate of leaf turnover and the withdrawal of nutrients from senescing blades (Josselyn et al. 1986, Hemminga et ale 1991). Periodic or continual shedding of leaves in some seagrass species may minimize the light inhibitory effects of epiphytes, upon seagrass photosynthesis (Sand-Jensen 1977, Borum and Wium-Anderson 1980, Bulthuis and Woelkerling 1983, Borowitzka and Lethbridge 1989). We lack enough basic information on the physiology and ecology of ~ hawaiiana to assign a beneficial or detrimental role to its epiphyte community. That basic knowledge should certainly precede speculation about seagrass growth dynamics as evolutionarily adaptive.

Establishment of seagrass and epiphyte associations In an early work designed to assess the basis for apparently specific seagrass and epiphyte associations, Harlin (1971) found that colonization on seagrass clones

19 (artificial sUbstrata) and the viable seagrass leaves was similar in composition. While some host and epiphyte associations appear to be specific, the specificity may be based on the seagrass habitat rather than the host surface

(Wood 1959, HUmID 1964, Wood 1972, Ballantine 1979, Harlin

1980). The reducing environment and the wide range of pH

(5.8 to 9.4) in sediments of seagrass meadows may exclude some species from beds (Wood 1959, 1972).

Few examples of exclusive associations between hosts and epiphytes occur and anatomical studies have shown physical connections between specific hosts and colonists

(Ducker and Knox 1984). Epiphytes termed "non-specific" show no special holdfast adaptations. These epiphytes are regarded as opportunists and are common in the littoral zone but do not generally colonize young hosts. Active defense mechanisms are thought to prevent their settlement on young leaf tissue (Ende and Linskens 1974). The settlement and attachment of opportunists is thought to be facilitated by the host aging process that may reduce the production of defense chemicals (Ducker and Knox 1984).

In marked contrast, there are epiphytes which are found only on seagrass leaves. Specialized holdfast morphologies and specific adhesives are involved in these selective associations. Recognition mechanisms, such as lectins, may allow specificity in the attachment of propagules to host

20 surfaces (Ducker and Knox 1984) and generate non-random patterns of recruitment (Borowitzka and Lethbridge 1989). The pool of propagules that might settle on a leaf determines potential colonists. The rapidity and strength of spore attachment may influence the ultimate composition of an epiphyte flora. Other features, such as spore diameter or weight, may facilitate their retention in the boundary layer of seagrass leaves (Boney 1975). Algal propagules could be produced via one of three modes: 1) vegetative cell division (asexual propagules of Sphacelaria), 2) thallus fragmentation (Polysiphonia and ceramium), or 3) spore formation (diploid or haploid and flagellated or non-flagellated). Many seagrasses grow as a "conveyer belt" with the youngest leaf or part of the leaf closest to the basal meristem and the oldest leaf part highest in the water column. New leaves are generally produced at nodes along an underground rhizome and formed at distinct time intervals. Thus, patterns in epiphyte communities on different aged leaves record changes in succession because epiphyte presence on leaves can be measured as a function of the age continuum in and among leaves (Novak 1984). Along this continuum, diversity and density of epiphytes typically increase from the youngest to the oldest leaf of a single plant, and from the youngest to the oldest area of a single leaf (Harlin 1980, Novak 1984, Brouns and

------Heijs 1986, Josselyn et ale 1986, Mazzella and Alberte 1986). Older leaves have been available for colonization the longest amount of time and physiological changes may occur with leaf age that could affect species composition (Novak 1984) and abundance (Cariello and Zanetti 1979a, Novak 1984, Brouns and Heijs 1986). Examples of these changes include a decline in the concentration of specific phenolic compounds in leaf tissue over time (Cariello and Zanetti 1979b) and differential release of dissolved organic carbon along the leaf axes (Penhale and Smith 1977, Velimirov et ale 1982, Kirchman et ale 1984, Pirc 1985). The leaves of Halophila are produced in pairs from the apical meristem of the rhizome. The oldest leaves are furthest from the meristem and occur at the same height in the water column as do young leaves. We have no descriptions of the fundamental chemistry and physiology for different aged leaves of Halophila hawaiiana. Colonization of new substrata is one of the most precarious stages in the life history of marine benthic organisms (Fletcher and Baier 1984). Highly specialized processes of settlement and early growth have developed to meet this "challenge" in the bacteria (Corpe 1970b, Corpe et ale 1976, Dempsey 1981), diatoms (Jones et ale 1983), macroalgae (Fletcher 1976, 1977, Jones et ale 1983) and invertebrates (Crisp 1973, Crisp et ale 1983). Substrate characteristics that may influence attachment and settlement

22 of propagules are addressed later in the section in the section on surface properties and modifying films. Three phases in the establishment of bacteria and diatom dominated communities are known, based on experiments with artificial substrata. First, bacteria settle; second, diatoms settle, clump and accumulate detritus; and finally, clumps overgrow each other, and the species number and density reach a maximum (Hudon and Bourget 1981). Overall epiphyte community morphology may be influenced by space limitation and structure in several ways. Diatoms that are initial colonizers occur mostly as prostrate forms (Korte and Blinn 1983), with occasional erect forms (Hudon and Bourget 1981, Ferreira and Seeliger 1985). The mucilaginous films of bacteria and diatoms are assumed to be significant in the attachment of other organisms (DiSalvo and Daniels 1975, Huang and Boney 1983). The films are thought to supply nutrients as well as modify surfaces for settlement (Huang and Boney 1983). Algal spores and animal larvae are attracted to the heterogeneous biofilms (Kirchman et ale 1984). Algal development may sustain bacterial productivity as the algal cells provide physical refuge, mucilage and growth-promoting substances (Stock and Ward 1989). The surface chemistry of epiphytic microbes may produce compounds, such as acidic polysaccharides (Corpe 1970a), that function in the adhesion of particulate materials to surfaces. Because of their surface charge and other

23 properties, these molecules act as polymer bridges between surfaces. The roles of specific compounds, forces and surface modifying processes, relevant to settlement on surfaces are discussed in the section on forces of attachment and the section on surface properties and modifying films. Microbial colonizers are sensitive to current direction and velocity (Korte and Blinn 1983). This is reflected as variations in species diversity, micro-patterns in biofilm topography, and rates of biofilm development. Distribution of algal epiphytes on the leaf (Harlin 1971), as well as the cell height and density, also vary as a function of water velocity (Korte and Blinn 1983). In one system, the initial rate of epiphyte colonization was inversely related to current velocity, while light energy accounted for differences in species presence subsequent to the initial colonization phase (steinman and McIntire 1986). Surfaces in the ocean have been considered neutral to colonists (Cattaneo and Kalff 1978, Blindow 1987). Alternatively highly selective behaviors have been demonstrated in some systems (Maki and Mitchell 1986). Underlying bases for patterns in incipient, and perhaps sUbsequent stages of community development, include the interfacial forces that drive attraction, adhesion, aggregation of cells and settlement. Additionally, cell surfaces are molecular mosaics that interact with their

------environment. In the next sections I discuss topics that have been identified to play roles in these fine-scale processes of colonization.

Forces of attachment "Supply-side ecology" (Roughgarden et al. 1986) would argue that the probability of propagules encountering a surface is influenced by their abundance in the water column. Particles including propagules or microorganisms are transported by fluid dynamic forces to the point where physico-chemical forces, such as electrostatic forces and van der Waals (electrodynamic) forces act upon cells near a surface (van Loosdrecht et al. 1989). A cell approaching a surface will first be exposed to these nonspecific interactions (Absolom et al. 1984). Only van der Waals forces operate at distances greater than approximately 50 nm. As the distance decreases (to approximately 10-20 nm) between the propagule and a surface, both van der Waals and electrostatic forces come into play. The potential energy barrier separating a cell and a surface is overcome at very short distances (1.5 nm or less), and a variety of short range specific or non-specific bonding can lead to irreversible adhesion (Busscher and Weerkamp 1987). This interplay of forces has been demonstrated in some plant/bacterial associations (van Loosdrecht et al. 1989).

25 Attachment depends on the ability of the colonizing entity to adhere initially to surfaces. A cell wall adheres by electrostatic (columbic) and van der Waals (dispersion) forces, surface charge and the wetability of the surface. Electrolytes (eg. Ca++) can overcome initial repulsive barriers and so facilitate initial attachment (Decho 1990). This initial process is reversible (Marshall et al. 1971) because the forces are relatively weak, and cells on the surface are sUbject to hydrodynamic shear forces prior to irreversible attachment (van Loosdrecht et al. 1989). It is not known if the initial attachment of bacteria to surfaces is an active or passive process. Several attachment mechanisms may be used depending on the substrate, local environment and particular bacterial species (Fletcher 1988a). The initial adhesion is sustained by particular attachment mechanism. Adhesive mucilages are secreted by microorganisms and propagules upon contact with surfaces (DiSalvo and Daniels 1975). Adhesive polymers from the cell bridge the distance of the aqueous layer between a cell and the surface (Decho 1990). Common ions in seawater, such as Ca++ and Mg++, may act as ionic bridges to cross-link adjacent sugars on separate exocellular polymeric sugars (Fletcher 1988b). Specific molecules within the exopolymer secretions (EPS) sustain attachment of microbes to surfaces (Costerton 1984, Tosteson 1985). These molecules include high molecular weight polymers of polysaccharides (Jones et

26 ala 1969, Fletcher and Floodgate 1973) and glycoproteins (Fletcher and Marshall 1982b). Other interactions in biological systems that affect attachment include: 1) hydrogen bonding (Pringle and Fletcher 1983, Fletcher 1988a), 2) hydrophobic bonding (Marshall and Cruickshank 1973, Pringle and Fletcher 1983), 3) coordination with metals or other cations, 4) polar group interactions, 5) steric interferences (Maroudas 1973, Maroudas 1975a,b), and, 6) specific reactions between the substrate and surface functional groups (Dazzo et al. 1976). Microbial adhesion to a substratum may result from: 1) specific surface structures, such as pili (fimbriae) or other appendages (Jones 1975),2) cell surface adhesives (Fletcher and Floodgate 1973),3) cell-surface plasticity

(Corpe 1970b), and, 4) hydrophobic areas of the cell surface (Marshall and Cruikshank 1973, Dempsey 1981). All these surface properties may be involved in epiphyte-host bonding. Physicochemical models for adhesion of bacteria to

substrata include: 1) electrokinetics and 2) thermodynamics (Fletcher 1988a). The electrokinetic model is based on charge interactions between the cell and substrate surfaces. At physiological pH most bacteria and potential substrata have a net negative charge. Electrostatic repulsion is proposed to occur between surfaces from an overlap between the electrical double layers of the charged groups on opposing surfaces. Adhesion depends upon overcoming these

27 repulsive forces. According to van Loosdrecht et ale (1989): This interaction depends on the surface potentials and the thickness of the electrical double layers. The thickness is inversely proportional to the square root of the ionic strength. At high electrolyte concentration or in the presence of polyvalent counterions, the electrostatic interaction will be reduced. While it is generally accepted that electrolyte concentrations in seawater are sufficient to overwhelm electrostatic repulsions between negatively charged bacteria and surfaces (Rutter and Vincent 1984), the adhesion of some microbes to charged surfaces have been observed (Fletcher and Loeb 1979). In contrast with the electrokinetic model, the thermodynamic model is based on the importance of hydrogen bonding, dipole and hydrophobic interactions (Pringle and Fletcher 1983). A relationship between free-energy of a solid's surface and bacterial colonization has been demonstrated (Fletcher and Loeb 1979, Absolom et ale 1984).

Surface properties and modifying films Surface characteristics of both substrates and cells play crucial roles in the initial stages of attachment and subsequent settlement (Baier 1973, Fletcher and Baier 1984). Important substrate characteristics that I discuss in the following sections are: 1) surface texture, shape and area, 2) surface tension, 3) surface hydrophobicity, and 4)

- ._---_._-- ._------surface charge. In addition to these interactions, surfaces are modified by macromolecular films and biofilms that alter initial surface characteristics.

Surface texture The rugosity of a substrate defines, in part, the microhabitat for any potential colonist. In general, the rougher or more uneven the solid surface, the more susceptible it is to the attachment of foreign molecules and micro-organisms (Zobell 1972). Adsorption rates will be higher for cells arriving at "microrough" surfaces and show a decreasing desorption rate as the absorbed cells are sheltered from shear forces, this increases the surface area for cell-substratum contact (Characklis 1990). Interstitial spaces may enhance settlement of spores in two ways. First, trapped organic constituents may give local electrical charges to the surface. Additionally, interstitial texture reduces local water velocities thereby increasing the residence time of spores in the local area (Harlin and Lindbergh 1977). Irregularly contoured surfaces typically support greater diversity of benthic organisms than do flat surfaces (Weise and Rhenheimer 1978, Hixon and Brostoff 1985). In general, spatial heterogeneity promotes co existence of species (Foster 1975, May 1986).

29 Surface Tension Linsken (1963) hypothesized that the surface tension of the host controls epiphyte settlement. While surface tension may be a barrier to settlement, certain marine organisms produce molecules that depress the surface tension of the substrata (Zobell 1972). Attachment and growth rates of bacteria on glass slides are affected by changes in surface tension (Linsken 1963). The critical surface tension of a substrate is determined by its molecular nature. Hard solids with high melting points have high critical surface tensions, e.g. silica and alumina. Soft solids are low energy materials, e.g. teflon. Low critical surface tension is more suitable for microbial adhesion. Cells are more evenly distributed in a discrete microlayer on surfaces with low critical surface tension, but differences among organisms have been observed (Fletcher and Loeb 1979). Distinctive patterns occur in the morphology and development of basal attachment systems in the green alga, Enteromorpha intestinalis, in response to changes in critical surface tension. Fletcher and Baier (1984) consider the ecological significance of this result: Successful establishment of an alga on a substratum will very much depend upon its adherent properties, particularly during the early more critical stages of settlement and growth. The ability of different critical surface tensions to influence the morphogenic development of the attachment rhizoids and thereby effectively control the strength of adhesion of the

30 plants will undoubtedly play a major determining role in the colonization of substrata by algae. These researchers hypothesized that critical surface tension may be a crucial factor influencing many host and epiphyte relationships.

Hydrophobic surface properties The lack of attraction between water and a surface is termed hydrophobicity. Surface hydrophobicity functions in adhesion by removal of the water film between interacting surfaces, enabling specific short-range interactions to occur (Busscher and Weerkamp 1987). Hydrophobicity may be the most important factor in the nonspecific adhesion of bacteria to interfaces (Fattom and Shilo 1984). Fattom and Shilo (1984) have found that the outer surface layers of benthic cyanobacteria are consistently hydrophobic while the surfaces of planktonic cyanobacteria are hydrophilic. What molecules impart a hydrophobic/hydrophilic character to cell surfaces? Cell wall proteins and lipids are capable of presenting hydrophobic areas by adopting particular conformations and orientations at the cell surface. Some polysaccharides present surfaces that are essentially apolar while others, e.g. cellulose, have fixed conformations with all the sugar hydroxyl groups forming a polar surface.

31 Surface charge Surface charge, in biological systems, functions in the transport of molecules and surface recognition as well as cell agglutination and adhesion (Martin 1987). Net charges on algal surfaces result from the relative concentrations of neutral and acidic polysaccharides (Barclay 1983). Charge interactions are involved in the early coating of a surface in the marine environment. Charged molecules (e.g. certain amino acids or sugars) as well as hydrophobic surface-active organic compounds (e.g. fatty acids and glycoproteins) are often the first to be concentrated on surfaces. Organic films facilitate attachment of bacteria to surfaces via an apparent change in the charge and free energy of the surface (Fletcher and Marshall 1982). Generally, particles in freshwater, estuarine and marine environments are negative; this charge is attributed to organic coatings. The charge of a substrate may be negative, positive, or neutral when compared with potential colonists; the net charge on a cell wall is a function of the concentration of positively and negatively charged groups in the wall components (Loder and Liss 1985). The metabolism of a host may influence its surface charge (Harlin 1980).

32 Surface modifications Surface coatings are ubiquitous in marine systems. They modify many chemical and physical surface properties including adsorption, charge (Characklis 1990), hydrophobicity and surface tension (Baier 1975). The duration of sUbmergence, the characteristics of the substrate, as well as the chemical, physical, and biological properties of the surrounding water determine the kinds and quantities of film-forming materials that attach to sUbmerged substrata (Zobell 1972). Baier (1973) proposed that the initial coating, the "conditioning film", is glycoprotein from the water column and that it functions as a surface for bonding with the mucopolysaccharides exuded from the first biotic colonizers. This conditioning film precedes attachment of microbes and the development of microbial biofilms. These biofilms consist of macromolecular, subcellular and microbial components. The bacterial glycocalyx forms the matrix of the biofilm. Microbes within biofilms have patchy distributions. This patchiness produces steep redox gradients within the biofilm, and promotes diffusive exchange of metabolites, gases and nutrients. Diverse microhabitats are created, that increase community diversity and stabilize "microzones" in the biofilm (Paerl 1982). A biofilm forms a metabolically active interface with colonists (Maki et ale 1990). Bacteria influence the

------sUbsequent adsorption of inert particles (DiSalvo and Daniels 1975), diatoms and other microorganisms, algal spores and animal larvae to substrates (Dempsey 1981). Extracellular material released from attached diatom cells has been shown to support associated bacterial activity and growth (Murray et al. 1986). Community development on a surface may be mediated by these microbial consortia as they metabolically interact and as their extracellular polysaccharides enhance microbial binding to the biofilm surface. The identification of microbial consortia in biofilms is an "interesting new development in microbial ecology" (White 1988). Microbial flora within certain biofilms exist as groups of physiological units rather than as a random assortment of species embedded within an exopolymer matrix (Costerton 1984).

Glycoprotein-mediated cell surface interactions In order for adhesion of microbes to occur, the cells must reside near the surface with adequate time for attachment and the cell surface must have a mechanism for recognition and attachment. The mechanism may involve lectins, a class of proteinaceous compounds that bind carbohydrates specifically and noncovalently. Lectins bond specifically with glycoconjugates (glycoproteins,

34 glycolipids and polysaccharides) on cell surfaces (see Sharon and Liss 1989 for a recent review). In the marine environment, glycoconjugates are present in biological exudates (Tosteson and zaidi 1974, Baier and Weiss 1975, Tosteson et ale 1976, Tosteson and Corpe 1975, Jimenez et ale 1979, Tosteson et ale 1985, Decho 1990). They contribute to the pool of components for the conditioning film that precedes settlement by microbes and biofilm development (Baier 1973). These macromolecules have an activity similar to that of lectins; they enhance both adhesion of microalgae to glass and aggregation to each other (Baier and weiss 1975, Tosteson and Corpe 1975, Tosteson and Zaidi 1974, Iman et ale 1984). Associations of bacteria and algal hosts may be due to selective cell interactions at the molecular level. In a study of an algal and bacterial interaction, Iman et ale (1984) conclude that both interspecific and intraspecific surface interactions were enhanced by macromolecules specific to the receptors on the cell surfaces. They further state that these processes may be of fundamental importance in the regulation of microbial surface interactions in the sea. Biochemical processes control the settlement of fouling organisms via interactions between bacterial films and marine invertebrates (Kirchman et ale 1982a). Observations that a larva can change its attachment response upon contact with specific bacterial films, and that a bacterial polymer

35 has both stimulatory and inhibitory components, suggest a complex relationship between the substrate, bacterium, and larva (Maki et ale 1990) or algal spore. Films generated by specific bacteria increase the recruitment of marine invertebrate larvae to surfaces (reviewed by Maki and Mitchell 1986). Bacterial films have also been shown to be important in the settlement and metamorphosis of invertebrate larvae (Crisp 1973). Settlement and metamorphosis of some larvae is viewed as a lectin-mediated surface interaction (Kirchman et ale 1982a) Discrete distributions of glycoconjugates on cell surfaces have been demonstrated for select species in a variety of algal divisions: Cyanophyta, Pyrrhophyta, Raphidophyta, Euglenophyta, Chromophyta and Chlorophyta (Vanni et ale 1981, Surek and Sengbusch 1981, Sengbusch et ale 1982, Sengbusch and Muller 1983, Iman et ale 1984, Kaska et ale 1988). In unicellular green algae, glycoconjugates are involved with cell-cell recognition, adhesion, morphogenesis and wall assembly (Iman et ale 1984, Klis et ale 1985, Goodenough et ale 1986, Schlipfenbacher et ale 1986, Samson et ale 1987). Attachment behavior, in some systems, has been demonstrated to involve lectins, or lectin-like molecules, and receptors at the cell surface (Iman et ale 1984). The localization of lectins and glycoconjugates, as potential adhesives, may further our understanding of basic processes

36 involved in the early stages of community development in the benthic marine environment.

Focus on Halophila hawaiiana as host for epiphytes As hosts, macrophytes differ from one another in morphology, anatomy, surface rugosity, and chemistry of both the photosynthetic surface and exuded products. How might our information on adhesion help explain patterns in benthic marine communities? Bacteria are generally considered to be the first colonists on substrates in the ocean. Bacteria interact with diatoms, form microbial consortia in a biofilm, and change the micro-topography and chemistry of the surface microzone. Settlement and metamorphosis of some marine organisms may require specific biofilms. Each colonist may have specific requirements for adhesion, and when attached alter the surface for subsequent colonists. Once a film has formed, the surface properties change and remain changed after the initial colonists disappear. Colonization and community development on living substrates involve a plethora of cellular interactions that form the bases for observable patterns. Knowledge of the growth dynamics and ecology of the seagrass, Halophila hawaiiana, as a host, sets the stage for observing both patterns in epiphyte community development and understanding the mechanisms that contribute to the patterns.

37 LITERATURE CITED

Absolom, D. R., Lamberti, F. V., Policova, z. zingg, W., van Oss, C. J., Neumann, A. W. (1984). Surface thermodynamics of bacterial adhesion. Appl. Environ. Microbiol. 46:90-97. Baier, R. E. (1973). Influence of initial surface conditioning of materials on bio-adhesion. In: Proc. 3rd into congr. on marine corrosion and fouling. Northwestern University Press, Ill. Baier, R. E. (1975). Applied chemistry at protein interfaces. In: Baier, R. E. (ed.) Applied chemistry at protein interfaces, advances in chemistry series, 145. American Chemical Society, Washington. Baier, R. E., Weiss, L. (1975). Demonstration of the involvement of adsorbed proteins in cell adhesion and cell growth. Adv. Chem. Sere 145:300-7. Ballantine, D. L. (1979). The distribution of algal epiphytes on macrophyte hosts offshore from La Parguera, Puerto Rico. Bot. Mar. 22:107-111. Barclay, W. R. (1983). Influence of nutrient limitation on the exocellular polysaccharide composition and resulting cell surface charge in Chlamydomonas sajao. J. Phycol. 19:2-15. Barnabas, A. D., Butler, V., Steinke, T. D. (1977). Zostera capensis Setchell. I. Observations on the fine structure of the leaf epidermis. Z. pflanzenphysiol. Bd. 85.S.:417-427. Barnabas, A. D., Butler, V., steinke, T. D. (1982). Zostera capensis Setchell. Some aspects of wall ingrowth development in leaf blade epidermal cells. Protoplasma 110:87-94. Birch, W. R., Birch, M. (1984). Succession and pattern of tropical intertidal seagrasses in Cockle Bay, Queensland, Australia: a decade of observations. Aquat. Bot. 19:343 367. Blindow, I. (1987). The composition and density of epiphytes on macrophytes--the neutral substrate hypothesis tested. Aquat. Bot. 29:157-168.

38 BUlthuis, D. A., Woelkerling, W. J. (1983). Biomass accumulation and shading effects of epiphytes on leaves of the seagrass Heterozostera tasmanica in Victoria, Australia. Aquat. Bot. 16:137-148. Boney, A. D. (1975). Mucilage sheaths of spores or red algae. J. Mar. BioI. Ass. U.K. 55:511-518. Borowitzka, M. A., Lethbridge, R. C. (1989). Seagrass epiphytes. In: Larkum, A. W. D., McComb, A. J., Shepherd, S. A. (eds.) Biology of seagrasses. Elsevier Sci. Pub. Co., New York. Borum, J. (1985). Development of epiphytic communities on eelgrass (Zostera marina) along a nutrient gradient in a Danish estuary. Mar. BioI. 87:211-218. Borum, J., Wium-Andersen, S. (1980). Biomass and production of epiphytes on eelgrass (Zostera marina L.) in the Oresund, Denmark. Ophelia, Suppl. 1:57-64. Borum, J., Kaas, H., Wium-Anderson, S. (1984). Biomass variation and autotrophic production of an epiphyte macrophyte community in a coastal Danish are. II. Epiphytic species composition, biomass and production. ophelia 23:165-179. Brawley, S. H., Adey, W. H. (1981). The effect of micrograzers on algal community structure in a coral reef microcosm. Mar. BioI. 61:167-177. Brawley, S. H., Xiugeng, F. (1987). Studies of mesoherbivory in aquaria and in an unbarricaded mariculture farm on the Chinese coast. J. Phycol. 23:614-623. Brouns, J. J. W. M., Heijs, F. M. L. (1986). Production and biomass of the seagrass Enhalus acoroides (L.f.) Royle and its epiphytes. Aquat. Bot. 25:21-45. Brown, C. S. (1962). On the ecology of aufwuchs of Zostera marina in Charlestown Pond, Rhode Island. M.A. thesis, University of Rhode Island. Kingston, R.I. Burkholder, P. R. (1973). The ecology of marine antibiotics and coral reefs. In: Jones, O. A., Endean, R. (eds.) Biology and geology of coral reefs 2, Biology 1. Academic Press, New York. Busscher, H. J., Weerkamp, A. H. (1987). Specific and non specific interactions in bacterial adhesion to solid substrata (review). Fems. Micr. 46(2):165-173. 39 Byrlinsky, M. (1977). Release of dissolved organic matter by some marine macrophytes. Mar. BioI. 39:213-220. cambridge, M. L., Chiffings, A. W., Brittan, C., Moore, L., McComb, A. J. (1986). The loss of seagrass in Cockburn Sound, Western Australia. II. Possible causes of seagrass decline. Aquat. Bot. 24:269-285. Cattaneo, A., Kalff, J. (1978). Seasonal changes in the epiphytic community of natural and artificial macrophytes in Lake Memphremagog (Que. and Vy.). Hydrobiologia 60:135-144. Capone, D. G. (1983). N2 fixation in seagrass communities. Mar. Tech. Soc. J. 17:32-37. Capone, D.G., Taylor, B.F. (1980). Microbial nitrogen cycling in a seagrass community. In: Kennedy, V. S. (ed.) Estuarine perspectives. Academic Press, New York. Cariello, L., Zanetti, L. (1979a). Effect of Posidonia oceanica extracts on the growth of Staphylococcus aureus. Bot. Mar. 22:129-131. Cariello, L., Zanetti, L. (1979b). Distribution of chicoric acid during leaf development of Posidonia oceanica. Bot. Mar. 22:359-360. Cariello, L., Zanetti, L., DeStefano, S. (1979). Posidonia ecosystem V. Phenolic compounds from marine phanerogams, Cymodocea nodosa and Posidonia oceanica. Compo Biochem. Physiol. 62 B:159-161. Chapman, A. R. o. (1979). Biology of seaweeds. University Park Press, Baltimore. Characklis, W. G. (1990). Biofilm processes. In: Characklis, W. G., Marshall, K. C. (eds.) Biofilms. John Wiley and Sons, Inc., New York. Cinelli, F., Cormachi, M., Furnari, G., Mazzella, L. (1984). Epiphytic macroflora of Posidonia oceanica (L.) Delile leaves around the Island of Ischia (Gulf of Naples). In: Boudouresque, C. F., de Grissac, J., Olivier, J. (eds.) International workshop on Posidonia oceanica beds. GIS Posidonie Publichers, Marseille.

Clements, F. E. (1916). Plant succession. An analysis of the development of vegetation. Carnegie Inst., Washington.

40 Clements, F. E. (1928). Plant succession and indicators. H.W. Wilson Co., New York. Connell, J. H. (1985). The consequences of variation in initial settlement vs. post-settlement mortality in rocky intertidal communities. J. Exp. Mar. BioI. Ecol. 93:11 46. Corpe, W. A. (1970a). An acid polysaccaride produced by primary film forming bacteria. Develop. Ind. Microbiol. 11:402-412. Corpe, W. A. (1970b). Attachment of marine bacteria to solid surfaces. In: Manly, R.S. (ed.) Adhesion in biological systems. Academic Press Inc., New York. Corpe, W. A., Matsuuchi, L., Armbruster, B. (1976). Secretion of Adhesive polymers of marine bacteria to surfaces. In: Proc. 3rd into biodegrade symp. Applied Science Publisher, London. costerton, J. W. (1984). Mechanisms of microbial adhesion to surfaces. Direct ultrastructural examination of adherent bacterial populations in natural and pathogenic ecosystems. In: Klug, M. J., Reddy, C. A. (eds.) Current perspectives in microbial ecology. proc. 3rd int. symp. Microbial Ecol. Crisp, D. J. (1973). Mechanisms of adhesion of fouling organisms. In: Acker, R. F., Floyd Brown, B., De Palma, J. R., Iveson, W. P. (eds.) Proc. 3rd int. congo on mar. corrosion and fouling. Northwestern University Press, Evanston. Crisp, D. J., Walker, G., Yule, A. B., Young, G. A., Woolmington, A. (1983). Settlement and adhesion of larvae-research at Menai Bridge. In: Mauchline, J. (ed.) Fouling and corrosion of metals in seawater. Scottish Marine Biological Association, Oban. D'Antonio, C. (1985). Epiphytes on the rocky intertidal red alga Rhodomela larix (Turner) C. Agardh: Negative effects on the host and food for herbivores? J. Exp. Mar. BioI. Ecol. 86:197:218. Dazzo, F. B., Napoli, C. A., Hubbell, D. H. (1976). Adsorption of bacteria to roots as related to host specificity in the Rhizobium-clover sYmbiosis. Appl. Environ. Microbiol. 32:166-171.

41 Decho, A. W. (1990). Microbial exopolymer secretions in ocean environments: their role(s) in food webs and marine processes. Oceanogr. Mar. BioI. Annu. Rev. 28:73-153.

Dempsey, M. J. (1981). Marine bacterial fouling-a scanning electron microscope study. Mar. BioI. 61(4):305-316. den Hartog, C. (1970). The seagrasses of the world. North Holland Publ., Amsterdam.

DiSalvo, L. H., Daniels, G. W. (1975). Observations on estuarine microfouling using the scanning electron microscope. Microb. Ecol. 2:234-240.

Doohan, M. E., Newcomb, E. H. (1976). Leaf ultrastructure and 13C values of three seagrasses from the Great Barrier Reef. Aust. J. Plant Physiol. 3:9-23.

Doty, M. S., Stone, B. C. (1966). Two new species of Halophila (Hydrocharitaceae). Brittonia 18:303-306.

Duarte, C. (1989). Temporal biomass variability and production/biomass relationships of seagrass communities. Mar. Ecol. Prog. Ser. 51:269-276.

Ducker, S. C., Knox, R. B. (1984). Epiphytism at the cellular level with special reference to algal epiphytes. In: Linskens, H.F., Heslop-Harrison, J. (eds.) Cellular interactions. Springer-Verlag, Berlin.

Ende, G. van den, Linskens, H. F. (1974). Cutinolytic enzymes in relation to pathogenesis. Ann. Rev. Phytopathol. 12:247-258.

Fattom, A., Shilo, M. (1984). Hydrophobicity as an adhesion mechanism of benthic cyanobacteria. Appl. Environ. Microbiol. 47:135-143.

Feldman, J. (1937). Recherches sur la vegetation marine de la mediteranee. La Cote des Alberes. Chap. 2. Le substratum. Revue Algol. 10:17-29.

Ferreira, S., Seeliger, U. (1985). The colonization process of algal epiphytes on Ruppia maritima L. Bot. Mar. 28:245-249.

Filion-Myklebust, C., Norton, T. (1981). Epidermis shedding in the brown seaweed Ascophyllum nodosum (L.) Le Jolis, and its ecological significance. Mar. BioI. Lett. 2:45 51.

42 Fitzgerald, G. P. (1969). Some factors in the competition or antagonism among bacteria, algae and aquatic weeds. J. Phycol. 5:351-359. Flahault, C., Schroter, C. (1910). Phytogeographische Nomenklatur. Proceed. Intern Bot. Congress, Brussels. Extrait du compte-rendu:1-29. Fletcher, M. (1976). The effects of proteins on bacterial attachment to polystyrene. J. Gen. Microbiol. 94:400 404. Fletcher, M. (1977). The effect of culture concentration and age, time, and temperature on bacterial attachment to polystyrene. Can. J. Microbiol. 23:1-6. Fletcher, M. (1988a). The significance of bacterial adhesion mechanisms in the formation of marine surface films. In: Thompson, M. F., Sarojini, R., Nagabhushanam, R. (eds.) Marine biodeterioration. A.A. Balkema, Rotterdam. Fletcher, M. (1988b). Effects of electrolytes on attachment of aquatic bacteria to solid surfaces. Estuaries 11(4):226-230. Fletcher, R. L., Baier, R. E. (1984). Influence of surface energy on the development of the green alga Enteromorpha. Mar. BioI. Lett. 5:251-254. Fletcher M., Floodgate, G. D. (1973). An electron microscopic demonstration of an acidic polysaccharide involved in the adhesion of a marine bacterium to solid surfaces. J. Gen. Microbiol. 74:325-334. Fletcher, M., Loeb, G. I. (1979). Influence of substratum characteristics on the attachment of a marine pseudomonad to solid surfaces. Appl. Envir. Microbiol. 37(1):67-72. Fletcher M., Marshall, K. C. (1982). A bubble contact angle method for evaluating substratum interfacial characteristics and its relevance to bacterial attachment. Appl. Environ. Microbiol. 44:184-192. Foster, M. S. (1975). Regulation of algal cummunity development in a Macrocystis pyrifera forest. Mar. BioI. 32:331-342. Fountaine, T. D., Nigh, D.G. (1983). Characteristics of epiphytic communities on natural and artificial submersed lotic plants. Archiv fur Hydrobiologie 96:293-301.

43 Gessner, F. (1968). Die Zellwand mariner Phanerogamen. Mar. BioI. 1:191-200. Gleason, H. A. (1926). The individualistic concept of the plant association. Bull. Torrey Botan. Club 53:7-26. Gleason, H. A. (1939). The individualistic concept of the plant association. Am. Midland Natutalist 21:92-110.

Goering, J. J., Parker, P. L. (1972). Nitrogen fixation by epiphytes on sea-grasses. Limnol. Oceanogr. 17(2):320 323.

Goff, L. J. (1982a). The biology of parasitic red algae. Prog. Phycol. Res. 1:298-369.

Goff, L. J. (1982b). Symbiosis and parasitism: another viewpoint. Biosci. 32: 255-256.

Goff, L. J., Coleman, A. W. (1984). Transfer of nuclei from a parasite to its host. Proc. Natl. Acad. Sci. 81:5420 5424. Goodenough, u. W., Gebhart, B., Meehan, R. P., Heuser, J. E. (1986). Crystal of Chlamydomonas reindardii cell wall: polymerization, depolymerization and purification of glycoprotein monomers. J. Cell BioI. 103:405-18. Gunning, B. E. S. (1977). Transfer cells and their roles in transport of solutes in plants. Sci. Prog. 64:539-568.

Gunning, B. E. S., Pate, J. S. (1969). "Transfer cells"- the plant cells with wall ingrowths, specialized in relation to short transport of solutes--their occurrence, structure and development. Protoplasma 68:107-133. Harlin, M. M. (1971). Epiphytic marine algae: interactions with their hosts. Ph.D. dissertation, University of Washington. Seattle, Wash. Harlin, M. M. (1973). Transfer of products between epiphytic marine algae and host plants. J. Phycol. 9:243-248. Harlin, M. M. 1975. Epiphyte-host relations in seagrass communities. Aquat. Bot. 1:125-131. Harlin, M. M. (1980). Seagrass epiphytes. In: Phillips, R.C., McRoy, C.P. (eds.) Handbook of seagrass biology. An ecosystem perspective. STPM Press, New York.

44 Harlin, M. M., Lindbergh, J. M. (1977). Selection of substrata by seaweeds: optimal surface relief. Mar. BioI. 40:33-40. Harrison, P. G. (1982). Control of microbial growth and of amphipod grazing by water-soluble compounds from leaves of Zostera marina. Mar. BioI. 67:225-230. Harrison, P. G., Chan, A. T. (1980). Inhibition of the growth of micro-algae and bacteria by extracts of eelgrass (Zostera marina) leaves. Mar. BioI. 61:21-26. Hemminga, M. A., Harrison, P. G., Lent, F. v. (1991). The balance of nutrient lossed and gains in seagrass meadows. Mar. Ecol. Prog. Sere 71:85-96. Herbert, D. A. (1984). The growth dynamics of Halophila hawaiiana, an endemic Hawaiian seagrass. M.A. thesis, University of Hawaii. Honolulu, Hawaii. Hixon, M. A., Brostoff, W. N. (1985). Substrate characteristics, fish grazing, and epibenthic reef assemblages off Hawaii. Bull. Mar. Sci. 37(1):200-213. Hootsmans, M. J. M., Vermaat, J. E. (1985). The effect of periphyton-grazing by three epifaunal species on the growth of Zostera marina L. under experimental conditions. Aquat. Bot. 22:83-88. Hough, R. A., Wetzel, R. G. (1975). The release of dissolved organic carbon derived from aquatic macrophytes: diel, seasonal, and community relationships. Bern. Int. Ver. Theor. Angew. Limnol. 19:939-948. Howard, R. K. (1982). Impact of feeding activities of epibenthic amphipods on surface fouling of eelgrass leaves. Aquat. Bot. 14:91-97. Howard, R. K., Short, F. T. (1986). Seagrass growth and survivorship under the influence of epiphyte grazers. Aquat. Bot. 24:287-302. Huang, R., Boney, A. D. (1983). Effects of diatom mucilage on the growth and morphology of marine algae. J. Exp. Mar. BioI. Ecol. 77:79-89. Hudon, C., Bourget, E. (1981). Initial colonization of artificial substrate: community development and structure studied by scanning electron microscopy. Can. J. Fish. Aquat. Sci. 38:1371-1384.

45 Humm, H. A. (1964). Epiphytes of the seagrass, Thalassia hemprichii, in Florida. Bull. Mar. Sci. Gulf Caribb. 14: 306-341. Iman, S. H., Bard, R. F., Tosteson, T. R. (1984). Specificity in marine microbial surface interactions. App. Envir. Micro. 48:833-9. Jagels, R. (1973). Studies of a marine grass, Thalassia testudinum I. Ultrastructure of the osmoregulatory leaf cells. Am. J. Bot. 60: 1003-1009. Jagels, R. (1983). Further evidence for osmoregulation in epidermal leaf cells of seagrasses. Am. J. Bot. 70:327 333. Jimenez, B. R., Tosteson, T. R., Zaidi, B. R., Tasi, T. S., Atwood, D. K. (1979). Adhesion enhancing organics in the Eastern Caribbean Sea. In: Stewart, Jr., H. B. (ed.) CICAR II Symposium of progress in marine research in the Caribbean and adjacent region. FAO Fisheries Report 200. Rome. Jones, A. M., Fletcher, R. L., Daniel, G. F., Jones, E. B. G. (1983). Settlement and adhesion of algae. In: Mauchline, J.(ed.) Fouling and corrosion of metals in seawater. Scottish Marine Biological Association, Oban. Jones, H. C., Roth, I. L., Sanders, w. M. (1969). Electron microscopic study of a slime. J. Bacteriol. 99:316-325. Jones, J. A. (1968). Primary productivity by the tropical marine turtle grass Thalassia testudinum, and its epiphytes. Ph.D. dissertation, University of Miami. Coral Gables, Fla. Jones, M. N. (1975). Biological interfaces: an introduction to the surface and colloid science of biochemical and biological systems. Elseviers, Amsterdam. Josselyn, M., Fonseca, M., Niesen, T., Larson, R. (1986). Biomass, production and decomposition of a deep water seagrass, Halophila decipiens Ostenf. Aquat. Bot. 25:47 61. Kaska, D. D., Polne-Fuller, M., Gibor, A. (1988). Developmental studies in Porphyra (Bangiophyceae). II. Characterization of the major lectin binding glycoproteins in differentiated regions of the Porphyra perforata thallus. J. Phycol. 24:102-7.

46 Kenworthy, W. J., Currin, C. A., Fonseca, M. S., Smith, G. (1989). Production, decomposition, and heterotrophic utilization of the seagrass Halophila decipiens in a submarine canyon. Mar. Ecol. Prog. Sere 51:277-290. Keough, M. J., Downes, B. J. (1982). Recruitment of marine invertebrates: the role of active larval choices and early mortality. Oecology 54:348-352. Kirchman, D., Graham, S., Reish, D., Mitchell, R. (1982a). Bacteria induce settlement and metamorphosis of Janua (Dexiospira) brasiliensis Grube (Polychaeta: Spirorbidae). J. Exp. Mar. BioI. Ecol. 56:153-163. Kirchman, D., Graham, S., Reish, D., Mitchell, R. (1982b). Lectins may mediate in the settlement and metamorphosis of Janua (Dexiospira) brasiliensis Grube (Polychaeta: Spirorbidae). Mar. BioI. Lett. 3:131-42. Kirchman, D. L., Mazzella, L., Alberte, R. S., Mitchell, R. (1984). Bacterial epiphyte on Zostera marina. Mar. Ecol. Prog. Sere 15:207-211. Klis, F. M., Samson, M. R., Touw, E., Musgrave, A., van den Ende, H. (1985). Sexual agglutination in the unicellular green algal Chlamydomonas eugametos. Identification and properties of the mating type plus agglutination factor. Plant Physiol. 159:740-746.

Klumpp, D. W., Howard, R. K., Pollard, D. A. (1989). Trophodynamics and nutritional ecology of seagrass communities. In: Larkum, A. W. D., McComb, A. J., Shepherd, S. A. (eds.) Biology of seagrasses. Elsevier Sci. Pub. Co., New York.

Korte, V. L" Blinn, D. W. (1983). Diatom colonization on artificial substrata in pool and riffle zones studied by light and scanning electron microscopy. J. Phycol. 19:332-342. KUo, J. (1978). Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae). I. Leaf blade and leaf sheath of Posidonia australis Hook F. Aquat. Bot. 5:171-190. KUo, J., Ridge, R. W., Lewis, S. V. (1990). The leaf internal morphology and ultrastructure of Zostera muelleri Irmisch ex Aschers. (Zosteraceae): a comparative study of the intertidal and subtidal forms. Aquat. Bot. 36:217-236.

47 Libes, M., Boudouresque, C. F. (1987). Uptake and long distance transport of carbon in the marine phanerogam Posidonia oceanica. Mar. Ecol. Prog. Sere 38:177-186. Linskens, H. F. (1963). Oberflachenspannung an marinen Algen. Koninkl. Nederl. Akade van Wetenschappen, Amsterdam. Reprinted from Proc., Series C 66:205-217. Loder, T. C., Liss, P. S. (1985). Control by organic coatings of the surface charge of estuarine suspended particles. Limnol. Oceanogr. 30:418-421. Lobban, C. S., Harrison, P. J., Duncan, M. J. (1985). The physiological ecology of seaweeds. Cambridge University Press, London. Maki, J. S., Rittschof, D., Samuelsson, M.-O., Szewzyk, U., Yule, A. B., Kjelleberg, S., Costlow, J. D., Mitchell, R. (1990). Effect of marine bacteria and their exopolymers on the attachment of barnacle cypris larvae. Bull. Mar. Sci. 46: 499-511. Maki, J. S., Mitchell, R. (1986). The function of lectins in interactions among marine bacteria, invertebrates, and algae. In: Mirelman, D. (ed.) Microbial lectins and agglutins. John Wiley and Sons, New York. Maroudas, N. G. (1973). Chemical and mechanical requirements for fibroblast adhesion. Nature 244:353 354. Maroudas, N. G. (1975a). Adhesion and spreading of cells on charged surfaces. J. Theor. BioI. 49:417:424. Maroudas, N. G. (1975b). Polymer exclussion, cell adhesion and membrane fusion. Nature 254:695-696. Marshall, K. C., stout, R., Mitchell, R. (1971). Mechanisms of the initial events in the sorption of marine bacteria to surfaces. J. Gen. Microbiol. 68:337-348. Marshall, K. C., Cruickshank, R. H. (1973). Cell surface hydrophobidity and the orientation of certain bacteria at interfaces. Arch. Microbiol. 91:29-40. Martin, R. E. (1987). Adhesion, morphology and locomotion of Paramoeba pemaguidensis Page (Paramoebidae): Effects of substrate charge density and external cations. J. Protozool. 34:345-348. May, R.M. (1986). The search for patterns in the balance of nature: advances and retreats. Ecology 67:1115-1l26. 48

------Mazzella, L. (1983). Studies on the epiphytic diatoms of Posidonia oceanica (L.) Delile leaves. Rapp. Camm. Int. Mer Medit. 28:127-128.

Mazzella, L., Alberte, R. S. (1986). Light adaptation and the role of autotrophic epiphytes in primary production of the temperate seagrass, Zostera marina L., J. Exp. Mar. BioI. Ecol. 100:165-180. Mazzella, L., Cinelli, F., ott, J. A., Klepal, W. (1981). studi sperimentali "in situ" sull' epifitismo della Posidonia oceanica Delile. Quad. Lab. Tecnol. Pesca 3:481-492.

McDermid, K. J. (1988). Community ecology of some intertidal subtropical algae, and the biology and taxonomy of Hawaiian Laurencia (Rhodophyta). Ph.D. dissertation, University of Hawaii. Honolulu, Hawaii.

McMillan, C. (1980). Culture methods. In: Phillips, R.C., McRoy C.P. (eds.) Handbook of seagrass biology. An ecosystem perspective. STPM Press, New York.

McMillan, C. (1984). The condensed tannins (proanthocyanidins) in seagrasses. Aquat. Bot. 20:351 357.

McMillan, C., Williams, S. C. (1980). Systematic implications of isozymes in Halophila section Halophila. Aquat. Bot. 9:21-31.

McMillan, C., Zapata, 0., Escobar, L. (1980). SUlphated phenolic compounds in seagrasses. Aquat. Bot. 8:267-278.

McRoy, C. P. (1974). Seagrass productivity: carbon uptake experiments in eelgrass, Zostera marina. Aquaculture 4:131-137.

McRoy, C. P., Barsdate, R. J. (1970). Phosphate absorption in eelgrass. Limnol. Oceanogr. 15:6-13.

McRoy, C. P., Barsdate, R. J., Nebert, M. (1972). Phosphorous cycling in an eelgrass (Zostera marina) ecosystem. Limnol. Oceanogr. 17:58-67.

McRoy, P., Goering, J. J. (1974). Nutrient transfer between the seagrass Zostera marina and its epiphytes. Nature 248:173-174.

McRoy, P., Goering, J.J., Chaney, B. (1973). Nitrogen fixation associated with seagrasses. Limnol. Oceanogr. 18:998-1002.

------Morgan, D., Kitting, C. L. (1984). Productivity and utilization of the seagrass Halodule wrightii and its attached epiphytes. Limnol. Oceanogr. 29:1066-1076. Moriarty, D. J. W., Boon, P. I., Hansen, J. A., Hung, W. G., Poiner, I. R., Pollard, P. C., Skyring, G. W., White, D. C. (1985). Microbial biomass and productivity in seagrass beds. Geomicrobiol. J. 4:21-51. Moriarty, D. J. W., Iverson, R. (1986). Exudation of organic carbon by the seagrass Halodule wrightii and its effect on bacteria growth in the sediment. J. Exp. Mar. BioI. Ecol. 96:115-126. Moss, B. L. (1982). The control of epiphytes by Halidrys siliguosa (L.) Syngb. (phaeophyta, Cystoseiraceae). Phycologia 21:185-188. Mueller-Dombois, D., Ellenberg, H. (1974). Aims and methods of vegetation ecology. John Wiley and Sons, New York. Murray, R. E., Cooksey, K. E., Priscu, J. C. (1986). Appl. Environ. Microbiol. 52:177-182. Murray, L., Wetzel, R. L. (1987). Oxygen production and consumption associated with the major autotrophic components in two temperate seagrass communities. Mar. Ecol. Prog. Sere 38:231-239. Novak, R. (1984). A study in ultra-ecology: microorganisms on the seagrass Posidonia oceanica. Mar. Ecol. 5:143 190. Orth, J. R., Van Montfrans, J. (1984). Epiphyte-seagrass relationships with and emphasis on the role of micrograzing: a review. Aquat. Bot. 18:43-69. Paerl, H.W. (1982). Chemical and biological alteration of the liquid-solid interface through bacterial microzone formation. meeting abstract. ABS PAP ACS 183:150. Palmer, M. A. (1988). Dispersal of marine meiofauna: A review and conceptual model explaining passive transport and active emergence with implications for recruitment. Mar. Ecol. Prog. Sere 48:81-91. Patriquin, D.G., Knowles, R. (1972). Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. BioI. 16:49 58.

50 Penhale, P. A. (1977). Macrophyte-epiphyte biomass and productivity in an eelgrass (Zostera marina L.) community. J. Exp. Mar. BioI. Ecol. 26:211-224. Penhale, P. A., smith Jr., W. O. (1977). Excretion of dissolved organic carbon by eelgrass (Zostera marina) and its epiphytes. Limnol. Oceanogr. 22:400-407. Penhale, P. A., Thayer, G. W. (1980). Uptake and transfer of carbon and phosphorous by eelgrass (Zostera marina L.) and its epiphytes. J. Exp. Mar. BioI. Ecol. 42:114-123. Peterson, C. H. (1986). Enhancement of Mercenaria mercenaria densities in seagrass beds: is pattern fixed during settlement season or altered by subsequent differential survival? Limnol. Oceanogr. 31:200-205. Pianka, E. R. (1983). Evolutionary ecology. Harper and Row, New York. Pirc, H. (1985). Growth dynamics in Posidonia oceanica. I. Seasonal changes of soluble carbohydrates, starch, free amino acids, nitrogen and organic anions in different parts of the plant. Mar. Ecol. 6:141-165. Pringle, J. H., Fletcher, M. (1983). Influence of substratum wettability on attachment of freshwater bacteria to solid surfaces. Appl. Environ. Microbiol. 45:811-817. Roughgarden, J., Gaines, S. D., Pacala, S. W. (1986). Supply side ecology: the role of physical transport processes. In: Gee, J. H. R., Giller, P. S. (eds.) Organization of communities: past and present. Blackwell Scientific Publications, Oxford. Rutter, P. R., Vincent, B. (1984). Physicochemical interactions of the substratum, microorganisms and the fluid phase. In: Marshall, K. C. (ed.) Microbial adhesion and aggregation. Springer Verlag, Berlin. Ryland, J. S. (1974). Observations on some epibionts of gulf-weed Sargassum natans (L.) Meyer. J. Exp. Mar. BioI. Ecol. 14:17-25. Sachet, M., Fosberg, F. R. (1973). Remarks on Halophila (Hydrocharitaceae). Taxon 22:439-443. Samson, M. R., Klis, F. M., Homan, W. L., van Egmond, P., Musgrave, A., van den Ende. H. (1987). Composition and properties of the sexual agglutinins of the flagellated green alga Chlamydomonas eugametos. Planta 170:314-22. 51 Sand-Jensen, K. (1977). Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot. 3:55-63.

Schlipfenbacher, R., Wenzel, S., Lottspeich, F., Sumper, M. (1986). An extremely hydroxyproline-rich glycoprotein is expressed in inverting Volvox embryos. FEBS Lett. 209:57-62.

Sengbusch, P. v., Mix, M., Wachholz, I., Manshard, E. (1982). FITC-Iabelled lectins and Calcofluor White ST as probes for the investigation of the molecular architecture of cell surfaces. Studies on conjugatophycean species. Protoplasma 111:38-52.

Sengbusch, P. v., Muller, U. (1983). Distribution of glycoconjugates at algal cell surfaces as monitored by FITC-conjugated lectins. Studies on selected species from Cyanophyta, Pyrrhophyta, Raphidophyta, Euglenophyta, Chromophyta and Chlorophyta. Protoplasma 114:103-13.

Sharon, N., Lis, H. (1989). Lectins as cell recognition molecules. Science 246:227-34.

Sieburth, J. M., Conover, J. T. (1965). Sargassum tannin, an antibiotic which retards fouling. Nature 208:52-53. smith, W.O., Penhale, P. A. (1980). The heterotrophic uptake of dissolved organic carbon in eelgrass (Zostera marina) and its epiphytes. J. Exp. Mar. BioI. Ecol. 48:233-242. steinman, A. D., McIntire, C. D. (1986). Effects of current velocity and light energy on the structure of periphyton assemblages in laboratory streams. J. Phycol. 22:352-361. stock, M. S., Ward, A. K. (1989). Establishment of a bedrock epilithic community in a small stream: Microbial (algal and bacterial) metabolism and physical structure. Can. J. Fish. Aquat. Sci. 46:1874-1883.

Surek, B., Sengbusch, P. v. (1981). The localization of galactosyl residues and lectin receptors in the mucilage and cell walls of Cosmocladium saxonicum (Desmidaceae) by means of fluorescent probes. Protoplasma 108:149-61.

Tansley, A. G. (1920). The classification of vegetation and the concepts of development. J. Ecol. 8:118-149.

Taylor, J. D., Lewis, M. S. (1979). The flora, fauna and sediments of the marine grass beds of Mahe, seychelles. J. Nat. Hist. 4:199-220. 52 Thayer, G. W., Adams, S. M., LaCroix, M. W. (1975). structural and functional aspects of a recently established Zostera marina community. Estuarine Res. 1:518-540.

Tomlinson, P. B. (1980). Leaf morphology and anatomy in seagrasses. In: Phillips, R. C., McRoy, C. P. (eds.) Handbook of seagrass biology: an ecosystem perspective. STPM Press, New York.

Tosteson, T. R., Atwood, D. K., Tsai, R. S. C. (1976). Surface active organics in the Caribbean Sea. In: Oceans 76 Symposium. Marine Technolgy Society. New York.

Tosteson, T. R., Corpe, W. A. (1975). Enhancement of adhesion of marine Chlorella vulgaris to glass surface. Can. J. Microbiol. 21:1025-31.

Tosteson, T. R., Revuelta, R., Zaidi, B. R., Iman, S. H., Bard, R. F. (1985). Aggregation-adhesion enhancing macromolecules and the specificity of marine microbial surface interactions. J. Colloid and Interface Sci. 104:60-71.

Tosteson, T. R., Zaidi, B. R. (1974). Surface active macromolecules in the marine environment: a sugar specific lectin-like activity. In: Weber, H. W., Ruggieri, G. D. (eds.) Food and drugs from the sea symposium. Marine Technol. Soc., Washington, D.C.

Tosteson, T. R. (1985). The regulation and specificity of marine microbial surface interactions. In: Colwell, R. R. (ed.) Biotechnology of marine polysaccharides. McGraw-Hill, New York.

Unabia, C. C. (1984). Grazing of the seagrass Halophila hawaiiana (Hydrocharitaceae) by the snail Smaragdia bryanae (Neritidae). M.A. thesis, University of Hawaii. Honolulu, Hawaii. van Loosdrecht, M. C. M., Lyklema, J., Norde, W., Zehnder, A. J. B. (1989). Bacterial adhesion: a physicochemical approach. Microb. Ecol. 17:1-15.

Vanni, G. L., Bonora, A., Dall'olio, G. (1981). Distribution of the receptors for Concanavalin A on the surface of Euglena gracilis as revealed by fluorescence microscopy. Plant Sci. Lett. 22:23-8.

Velimirov, B., ott, J. A., Novak, R. (1982). Microorganisms on macrophyte debris: Biodegradation and its implication in the food-web. Kiel. Meeresforsch Sonderh. 5:333-344. 53 Waldron, K. W. Baydoun, E. A.-H., Brett, C. T. (1989). Comparison of cell wall composition of tissues from the seagrasses Halophila and Halodule. Aquat. Bot. 35:209 218.

Walter, H. (1964). Die Vegetation der Erde in oko physiolischer Betrachtung. Band I: Die tropischen und sUbtropischen Zonen. 2nd ed., VEB Fischer, Jena. Walter, H. (1971). Ecology of tropical and sUbtropical vegetation. In: Burnett, J.H. (ed.). Oliver & Boyd, Edinburgh.

Weise, W., Rheinheimer, G. (1978). Scanning electron microscopy and epifluorescence investigation of bacterial colonization of marine sand sediments. Microb. Ecol. 4:175-188.

Wetzel, R. G., Penhale, P. A. (1979). Transport of carbon and excretion of dissolved organic carbon by leaves and roots/rhizomes in seagrasses and their epiphytes. Aquat. Bot. 6:149-158. Wetzel, R. G. (1983). Attached algal-substrata interactions: fact or myth, and when and how? In: Wetzel, R. G. (ed.) Periphyton of freshwater ecosystems. Dr W. Junk Publishers, The Hague. White, D. C. (1988). Assessment of marine biofilm formation, succession, and metabolic activity. In: Thompson, M-F., Sarojini, R., Nagabhushanam, R. (eds.) Marine biodeteriorationi advanced techniques applicable to the Indian Ocean. A.A. Balkema, Rotterdam. Wilson, F. S. (1990). Temporal and spatial patterns of settlement: a field study of molluscs in Bogue Sound, North Carolina. J. Exp. Mar. BioI. Ecol. 139:201-220.

Wood, E. J. F. (1959). Some aspects of the ecology of Lake Macquarie, N.S.W., with regard to an alleged depletion of fish. VI. Plant communities and their significance. Austr. J. Mar. Freshwater Res. 10:322-344.

Wood, E. J. F. (1972). Substratum: unicellular plants. In: Kinne, o. (ed.) Marine ecology, 1. Biologische Anstalt Helgoland, Hamburg. Wood, E. J. F., Odum, W. E., Zieman, J. C. (1969). Influence of seagrasses on the productivity of coastal lagoons. In: Coastal lagoons, a symposium, UNAM-UNESCO. univ. Nacional Autonomia Mexico, Mexico, D.F.

54 Zapata, 0., McMillan, C. (1979). Phenolic acids in seagrasses. Aquat. Bot. 7:307-317.

Zieman, J. C., Wetzel, R. G. (1980). Productivity in seagrasses: methods and rates. In: Phillips, R. C., McRoy, C. P. (eds.) Handbook of seagrass biology. STPM Press, New York.

Zobell, C. E. (1972). Bacteria, fungi and blue-green algae. In: Kinne, o. (ed.) Marine ecology: a comprehensive, integrated treatise on life in oceans and coastal waters. Environmental factors 3. Wiley-Interscience, London.

55 CHAPTER 2

LEAF ULTRASTRUCTURE OF THE HAWAIIAN SEAGRASS HALOPHILA HAWAIIANA Doty and stone

Abstract

The leaf ultrastructure of the endemic Hawaiian

seagrass, Halophila hawaiiana Doty and stone is documented

in this study. The seagrass epidermis differs from that of a terrestrial angiosperm. A thin cuticle covers the outer

epidermal cell walls. These outermost cells have thick cell walls and show distinctive convoluted regions of plasmalemma

and cell wall. These cells resemble transfer cells as described for some other angiosperms. The putatuive

transfer cell configuration is most elaborately developed in

the upper leaf surfaces. Chloroplasts, mitochondria,

endoplasmic reticulum, and dictyosomes are commonly

associated with ingrowth regions. Plasmodesmata occur

within epidermal, ground or vascular tissues. The only

symplastic intercellular connections between tissues occur

at the leaf margins. Intercellular spaces are common

between the epidermal and ground tissues and within the

ground tissue. Parenchyma cells within the ground tissue

are generally thin walled with large vacuoles and few

organelles. Chloroplast morphology varies with tissue type

and proximity to veins. Fine-structure differences are

apparent between the central, transverse and marginal veins.

The present study provides the first description of the fine

56 structure of this species providing a basis for interpretation of the function of the leaves and extends information applicable to a greater understanding of the genus Halophila.

Introduction The seagrasses are termed secondary aquatics because of their evolutionary origin in the terrestrial environment. While distinctive from both the primary aquatics (algae) and terrestrial angiosperms their form and function may hold similarities with both their environmental neighbors and evolutionary progenitors. systematics Seagrasses are monocotyledonous vascular plants that establish, grow, and propagate by both vegetative means and flowers while completely submerged in the ocean. They are an ecological assemblage, probably a polyphyletic group of angiosperms that evolved from one or more terrestrial progenitors (Dahlgren 1985, Tomlinson 1982). As a natural assemblage the seagrasses show a "remarkable array" (Tomlinson 1982) of morphologically specialized features. Common vegetative attributes of seagrasses are 1) the herbaceous habit, 2) simple leaves, 3) protoxylem lacunae, 4) absence of vessels, 5) hair-like scales in the leaf axis termed "squamules" in many species and 6) bifurcation of

57 apical shoots. Morphological architecture varies between species but is precise and highly organized for a given species (Dahlgren 1985). Halophila hawaiiana is a member of the superorder

Alismatiflorae (Helobae according to Tomlinson 1982), the order Alismatales and the family Hydrocharitaceae (Dahlgren

1985). Characteristics of the vegetative morphology for the super order include presence of squamules and vessels restricted to the roots if present at all. Anatomy of H. hawaiiana illustrates many vegetative characteristics for the order as a rooted, rhizomatous, aerenchymous marine aquatic. Shoots differentiate into petiole and lamina, and transverse veins converge at the apex. The family Hydrocharitaceae is a highly specialized family of herbaceous monocotyledons that are usually aquatic with either facultatively or obligately sUbmerged leaves. This family differs from others in the Alismatales in floral characteristics including inferior ovary, seeds without

endosperm and dissimilar fruit (Dahlgren 1985). Three marine genera of this family are Enhalus, Thalassia and Halophila. Halophila differs from the other genera in its lack of tannin cells, pinnate venation and epigyny. These morphological characteristics form a basis for the delineation of a monogeneric subfamily, Halophiloidae according to some systematic schemes (den

Hartog 1970, Dahlgren 1985).

58 The genus Halophila is described as a diminutive, rhizomatous marine herb (Tomlinson 1980) that is widely distributed in warm seas (Sachet and Fosberg 1973). Leaves and unbranched roots form at nodes on the rhizome and are separated by elongate internodes. Rhizome organization is considered a constant feature of the genus while the organization of erect shoots is variable. Subdivision of the genus is based on this organization of the erect shoots, flower position (Posluszny and Tomlinson 1991), the presence/absence of a ringlike annulus in the outer epidermal wall (Solreder 1913, Birch 1974), vein and leaf morphology (Doty and Stone 1966) and isozyme polymorphisms (McMillan and Williams 1980). Epidermal cells in all members of the genus are biseriate except in the vein areas. The ground tissue includes parenchyma cells interrupted by intercellular spaces and larger air canals. Vascular tissue occurs in the marginal, mid and transverse veins. The transverse veins ascend from the midrib, fork rarely and number 10 to 16 in the Hawaiian species (Doty and Stone 1966) Halophila hawaiiana Doty and Stone (1966) is the sole seagrass species in the Hawaiian Islands. Although still classified as a sUbspecies of H. ovalis by den Hartog (1970), the standing of H. hawaiiana has been generally respected. The Hawaiian species was separated from R. ovalis by Doty and Stone (1966) on the basis of spacing and

59 degree of branching of lateral veins from the midvein, smaller blades and fewer ovules. Sachet and Fosberg (1973) upheld separation of the Hawaiian species on the basis of ovule number but consider the leaf characteristics intermediate between H. minor and fl. ovalis. McMillan and Williams (1980) substantiated the genetic basis for separating R. hawaiiana based on its unique isozymes. While R. hawaiiana contains sulphated phenolic acids common to most seagrasses, it lacks the flavone sUlphates that occur in the R. ovalis complex and many other seagrasses.

Anatomy The morphology and anatomy of a plant is considered by some to be clearest expression of its adaptive success; ultrastructural studies form a basis for further investigations into metabolic processes that may be determinants in population structure, biotic and abiotic environmental factors (Tomlinson 1982). A number of seagrass ultrastructural studies have focused on the epidermis (Jagels 1973, 1983, Benedict and Scott 1976, Doohan and Newcomb 1976, Barnabas et ale 1977, 1980, 1982, Barnabas 1982, 1988, Colombo et ale 1983). In contrast with the epidermis of terrestrial plants, the outer seagrass cells contain abundant chloroplasts, mitochondria and secretory organelles that are spatially close to cell wall ingrowths and invaginated plasmalemma. The cell wall

60 ingrowths with invaginated plasmalemma have been interpreted as a basic module that is highly developed on the faces of the cell presumed to be most active in short distance solute transport. The ingrowths have been considered "the definitive features of transfer cells" (Gunning and Pate 1969). The overall geometry of plant tissue that is efficient in absorptive or secretory functions may be explained, in part, by the principle of "maximum exposure of surface" (Haberlandt 1914). The cell wall invaginations which increase the surface area of the plasmalemma have been implicated in these processes (Gunning and Pate 1969, Gunning 1977, Kuo et ale 1990). This arrangement occurs in most of the seagrasses that have been studied. The specific development and distribution of the cell wall elaborations is consistent within a taxon but differs between individual plants. Numerous mitochondria, chloroplasts and secretory organelles are consistently associated with the cell wall elaborations. Perhaps the organelle, plasmalemma and cell wall ingrowth form a functional module in the seagrass epidermis. Overall chemistry of the seagrass surface may affect and/or be a product of the metabolic processes occurring at the interface. Seagrass cell walls are thought to have certain ion exchange properties (Barnabas 1988) because of pectin as well as cellulose in the outer periclinal wall

--_. _._------(Gessner 1968, Kuo 1978, Kuo et ale 1990, Colombo et ale 1983). Halophila cell walls also contain uronic acids that may function in ion-exchange as do many of the acidic polysaccharides in the seaweeds (Chapman 1979, Waldron et ale 1990). Sulfated phenolic compounds in Halophila may play a role in osmotic adjustment to the marine environment as well as in the allelochemical relations of seagrasses (McMillan et ale 1980). The epidermal wall may be important in strengthening the leaf (Birch 1974). The internal leaf tissues function primarily in support, aeration and intercellular interactions via apoplastic, symplastic systems. Evidence for an apoplastic continuum from the epidermis through the mesophyll has been demonstrated for one seagrass, Thalassodendron ciliatum (Forssk.) Den Hartog (Barnabas 1988). Actual roles of the cell walls, intercellular spaces, large air canals, xylem (components of the apoplast) or plasmodesmata and phloem (components of the symplast) in the movement of solutes, solvents and intercellular communication have not been determined (Raven 1984). While the seagrass epidermis is more structurally complex than that of terrestrial angiosperms, the seagrass vascular tissue is generally reduced when compared with terrestrial angiosperms. Mature seagrass leaves can lack xylem or have protoxylem lacunae that form developmentally by stretching and enlargement during organ elongation (Esau

62 1977). Photosynthate is supplied to roots, rhizomes and propagules by phloem. This is the dominant cell type in each seagrass vascular system studied. Aerenchyma, large air canals and intercellular spaces, allow for intraplant gas exchange and internal support in each seagrass organ. Air canals develop from enlargement and expansion of existing intercellular spaces (Tomlinson 1982). The objective of this study was to document the ultrastructure of Halophila hawaiiana leaves. Application of this information will both extend our understanding of the genus and species, and provide a basis for investigating structural adaptations to specific environmental parameters.

Materials and Methods collection of Halophila hawaiiana Two populations of Halophila hawaiiana in La'ie Bay, O'ahu, Hawai'i was sampled during October, 1991. The populations were from wave exposed and wave sheltered sites. The wave exposed site is ca. 60 meters off the shore of Malaekahana beach. From approximately September through May this site is affected by a long shore current and shore break. The wave sheltered site is 10 meters off a small island, Moku'auia, ca. 800 meters north the Malaekahana beach site. This site is protected from waves throughout the year. Temperature and salinity were 26 0 C and 33 ppt in both sites at the time of collection.

63 The specific growth rates during that time period were determined by tagging replicate terminal apices and monitoring internode elongation and leaf production over time. Surveyors tape in 0.3 cm X 4.0 cm strips was tied around the rhizome one cm behind the terminal bud. The work was accomplished with the use of SCUBA. Growth was measured as that increment from the apex to the tape, leaf pairs were counted. The plastochrone interval, or the time occurring between initiation of sequential leaves (Tomlinson 1974) was 4 days during this time. Whole plants were collected for anatomical study; only mature (12 days old) unepiphytized leaves (less than 5% surface cover) were examined in this study. The seagrasses were transported to the lab in 0.4 ~m filtered seawater. Water temperature and salinity at the time of collection were 26° C and 33 ppt respectively.

Sample preparation for transmission electron microscopy Pre-senescent but mature, green leaves of the same age were fixed for transmission electron microscopy (TEM) by the following process: leaves were cut to ca. 1 to 2 rom transverse sections; sections underwent primary fixation in 2% glutaraldehyde buffered with 0.1 M sodium cacodylate in filtered seawater, Ph 7.6 for three hours. Ruthenium red at a concentration of 1% (Beveridge 1989) was added to the glutaraldehyde during this prefixation step to enhance

64 visualization of polysaccharides and pectic substances (Luft 1971, Blanquet 1976). Tissues were rinsed for 30 min. with buffered seawater and post-fixed in sodium cacodylate/seawater buffered 1% OS04 for one hour at room temperature. The tissue was then rinsed in buffered seawater, desalted in a graded series of buffered seawater to deionized water, dehydrated in a graded series of ETOH and finally embedded in Epon resin. Ultrathin sections of ca. 90 nm were cut with a diamond knife mounted in a Reinhart-Jung Ultracut E ultramicrotome, and then stained with saturated aqueous uranyl acetate and lead citrate (Reynolds 1963). Sections were mounted on Formvar-coated, single-slot grids and examined with a Zeiss 10/CA TEM, Reinhart-Jung, at 80 kv. The samples were examined in transverse section.

Observations with light microscopy Fresh, transverse sections were made free hand and stained for suberin with Sudan III and IV (Johansen 1940). Observations were made with a Zeiss Photomicroscope II. The leaf was viewed in face view with an Olympus BHS microscope and photographed with an Olympus model PM-lOADS automatic photomicrographic system.

. ---_..•.... _._.. _-----_._--- Results Ultrastructural details of the seagrass anatomy were similar in same-aged leaves from both wave exposed and sheltered sites. Each result reported was substantiated by repeated observations of replicate leaves. Halophila hawaiiana is rhizomatous and monopodially branched. SUb-opposite scale leaves, petiolate foliage leaf pairs and unbranched roots occur at regularly spaced nodes (Fig. 2.1). The spacing of one internode corresponds to one plastochrone interval. The plastochrone interval was found to be 4 days during the collection period. New foliage and scale leaves emerge in pairs from the terminal apex of the rhizome. Mature leaves from the node were calculated to be 12 days in age. Venation of the obovate foliage leaves is pinnate and includes a central vein, transverse veins oblique to the mid-vein and a marginal vein (Fig. 2.2). The veins connect at the leaf apex. While the leaf is generally isolateral, the abaxial (lower) leaf surface contains a greater number of epidermal cells and larger parenchyma cells around the central vein. This results in a slight asymmetry at the leaf center (Fig. 2.3). The facial view of the leaf (Fig. 2.4) shows a pattern defined by the anticlinal walls of the epidermis; cells between veins are isodiametric while those over the veins are elongate. The overall leaf surface is smooth.

66 The general construction of this leaf is shown in a light micrograph (Fig. 2.3). A leaf is formed of biseriate epidermal cells except in the three ground tissue-rich areas: the marginal, transverse and central veins. Distinct air canals occur in the midrib area. Intercellular spaces adjoin the epidermis to the ground tissue. Vascular tissue and air canals are bordered by thin walled parenchyma cells with scant peripheral cytoplasm and large vacuoles. Numerous, large chloroplasts are located in the densely cytoplasmic epidermal cells.

The epidermis

The upper and lower leaf epidermis are similar in having a thin cuticle (Fig. 2.5), thickened outer periclinal walls and an organelle-rich cytoplasm. The cuticle is electron translucent and slightly structured due to small striations or pores (Figs. 2.5,2.6). The organelles are generally peripheral in the cell and include: a large peripheral nucleus (Fig. 2.6), abundant chloroplasts and mitochondria (Figs. 2.5, 2.6). The chloroplasts have a dense stroma, lipid and stacked grana (Figs. 2.5, 2.6,

2.30). The mitochondria have well developed cristae and are closely associated with cell wall ingrowths and chloroplasts

(Figs. 2.5-2.7).

The outer cell walls show two zones as defined by electron density and fibril compaction. The outer wall has

67 compact fibrils and appears homogeneous. This outer wall is continuous with the middle of the anticlinal wall and appears similar then in electron density to the pectin-rich middle lamellae, or middle wall (Figs. 2.10, 2.11) and to some internal cells (eg. Figs. 2.16, 24). The fibrils of the inner wall are comparatively less compact (Figs. 2.5, 2.6) and have a repeating or "herring bone" pattern (Chafe 1974) when viewed at higher magnification (Figs. 2.7, 2.9). Distinctive cell wall ingrowths project into the epidermal cells from the inner surface of the outer periclinal cell walls (Figs. 2.5, 2.6). The cell wall ingrowths appear as papillae, branched in all directions. In both longitudinal and transverse section, these structures appear fibrillar (Figs. 2.5, 2.7). The cell wall ingrowths show a positive reaction to pectin as probed with the fixation component, Ruthenium red (Figs. 2.5-2.7, 2.9). The plasmalemma follows the contours of the cell wall ingrowths and is separate from the wall ingrowths by extracytoplasmic spaces (Figs. 2.5-2.7, 2.9). Endoplasmic reticulum, ribosomes, microtubules (Fig. 2.8) and dictyosome vesicles (Fig. 2.9) cluster in areas of the inner epidermal wall between cell wall ingrowths. Cell wall ingrowths with invaginated plasmalemma develop only on the inner surface of the outer epidermal cell wall. These structures are larger, longer and the most elaborately branched in the adaxial epidermis (contrast

68 Figs. 2.5 and 2.6). The associated chloroplasts of the adaxial and abaxial epidermis show differences in the grana that may relate to position in the leaf (contrast Figs. 2.5 and 2.6. Epidermal cells from both surfaces are connected by aggregates of five or more simple plasmodesmata (Figs. 2.10 2.12). Connections between epidermal cells at the leaf margin are more abundant, ranging from 7 to 20 plasmodesmata per aggregate. No symplastic connections were found between epidermal and ground tissue cells except at the leaf margins. Simple plasmodesmata (ca. five per cluster) adjoin the marginal epidermis to underlying parenchyma cells near the vein (Fig. 2.23). Epidermal and ground tissues are connected apoplastically via intercellular spaces (Fig. 2.23).

Ground tissue Ground tissue is developed in the vein areas. outermost parenchyma cells of the ground tissue differ from epidermal cells in that they are thin walled, the cytoplasm is less dense and a large central vacuole appresses the few organelles to the cell periphery (Fig. 2.13). Chloroplasts nearest to the epidermis contain lipid yet have less distinctly developed grana than epidermal cell chloroplasts (compare Figs. 2.30 and 2.31). Parenchyma cells adjacent to transverse veins contain large, central chloroplasts with

69 starch as well as lipid (Fig. 2.32). with the exception of these chloroplasts, organelles diminish in abundance with proximity to the vascular tissue. Parenchyma cells adjacent to both the vascular cells and large air canals are nearly devoid of cytoplasm (Figs. 2.16, 2.24, 2.26, 2.29). Internal parenchyma cells interconnect symplastically by simple plasmodesmata in aggregates of ca. three (Figs. 2.14,2.15). An apoplastic continuum involves intercellular spaces that occur between the ground tissue and epidermis, within the ground tissue and between the ground and vascular tissues (Figs. 2.13, 2.24, 2.26, 2.28, 2.29) as well as by contiguous the cell walls throughout the leaf.

Vascular tissue Vascular tissue occurs in transverse (Figs. 2.16-2.22), central (Figs. 24-26) and marginal veins (Fig. 2.23). Most cells lack nuclei and are highly vacuolate. Cells in the bundle sheath position, or the periphery of the vascular tissue, appear generally undifferentiated. The most abundant tissue in each vein is phloem. Sieve tUbes, companion cells, vascular parenchyma, and protoxylem occur within the vascular tissue of the leaf but not identically in each of the three veins. The most highly differentiated phloem occurs in the transverse veins (Figs. 2.16-2.22). Sieve tubes are distinguished by thickened walls that are less electron

70 dense than the peripheral phloem and contain sieve plates.

Cuneate protein bodies from sieve tUbe plastids (type PIIc,

Behnke 1981) and empty plastid membranes are evident in sieve tUbe cells (Figs. 2.16-2.18,2.21). Sieve pores in simple and compound sieve plates connect sieve tubes (Figs.

2.17, 2.18, 2.21). Plasma membranes are continuous through and around the pores (Fig. 2.18). The striated material within the pore space may be filaments of P-protein (Fig. 2.18) •

Plasmodesmata that adjoin phloem elements are evident in face (Figs. 2.18, 2.19) and transverse (Figs. 2.19-2.22) planes of view. Both simple and branched plasmodesmata occur between sieve tubes and companion cells (Figs. 2.19

2.21). The branches face the companion cell which contains microbodies, endoplasmic reticulum and ribosomes, vacuoles and secretory vesicles (Figs. 2.19-2.21). Callose deposition within the matrix of the primary wall increases the area of the plasmodesmata (Fig. 2.20-2.22). Both callose and desmotubules were observed in some branched plasmodesmata (Figs. 2.20, 2.22).

The transverse veins are bordered by large, vacuolate parenchyma cells (Figs. 2.16,2.17). Some cells in the bundle sheath position, contain distinctive chloroplasts that fill most of the cell volume and contain starch and protein-rich inclusions (Figs. 2.16, 2.33). These

inclusions resemble the sieve tUbe plastids described for

------Hydrocharitaceae (Behnke 1981). The vascular cells number ca. 17-20 per vascular bundle and are composed of phloem parenchyma with ca. 7-10 sieve tubes and companion cells.

Bundle sheath cells are similar to vascular parenchyma and are distinguishable only by position and the occasional chloroplasts with starch and protein inclusions (Figs. 2.16,

2.33). Air lacunae were not directly associated with the vascular tissue of the transverse veins.

The marginal veins are loosely bordered by thin walled parenchyma cells that underlay the epidermis. The vascular cells number ca. 10 to 13 per vascular bundle and are composed of less differentiated cells than the other veins.

Scattered chloroplasts (1 or 2) on some of the cells. Sieve tube cells (2 or 3) are evident. Phloem in the marginal vein is not bordered by bundle sheath cells. Both intercellular spaces and plasmodesmata are common to the epidermal and ground tissue cells that adjoin the vascular cells. Intercellular connections in this area are frequent.

The central vein is separated from the epidermis by 2 to 5 tiers of large, vacuolate parenchyma cells. These cells are more abundant in the lower surface and contribute to a thickening of this area of the midvein. The vascular cells number ca. 18 to 22 and are composed mainly of vascular parenchyma with ca. 6 to 8 sieve tubes per central vascular bundle. Sieve pore formation was observed at the stage of callose deposition (Figs. 2.24, 2.25). companion

- ._--_.. ------cells were not observed. As in the transverse veins, bundle sheath cells are mainly distinguishable by position. Protoxylem was observed only in the central vein (Figs. 2.24-2.26) adjacent to the parenchyma cells lining a large air canal. Wax as well as an electron-dense compound occur in the protoxylem wall (Fig. 2.27). Xylem was rarely observed in H. hawaiiana. Large air lacunae are a conspicuous feature of the central vein area. Two lacunae occur on each side of the central bundle (Fig. 2.3) and each is lined with vacuolate parenchyma cells that have thickenings along the walls facing the cavity lumen (Fig. 2.26). Intercellular spaces are common within the ground tissue near the central vein (Figs. 2.24, 2.26). Intercellular space formation may be by detachment of a discrete, pectin-rich plug (Fig. 2.25) and by cell wall degradation (Fig. 2.26). Results of the simple histochemistry that was carried out indicate that suberin is present in both the epidermis and the bundle sheath cells.

Discussion This study provides documentation of the fine-structure of leaves from the Hawaiian seagrass, H. hawaiiana. Ultrastructural studies of a closely related species, R. ovalis, have provided information on the "unusual epidermis" (Birch 1974) and the organelle-rich small structures,

73 squamules, at the leaf bases (Naidoo et al. 1990). Halophila hawaiiana and R. ovalis both have elaborate cell wall ingrowths inside the outer epidermal cell walls. These species differ in distinctive cell wall characteristics. The ingrowths are most elaborately developed in epidermal cells of the upper leaf surface in R. hawaiiana. Characteristics reported for H. ovalis but unreported for R. hawaiiana are a "collenchymatous" outer cell wall and an annulus in the epidermal cell wall with localized ion concentrations (Solereder 1913, Birch 1974). Additional fine-structural features that were determined for R. hawaiiana may be common in other members of the genus but have not been reported. These aspects include: 1) abundant and elaborate cell wall ingrowths on the inner surface of the outer epidermal periclinal cell walls, 2) plasmodesmata between adjacent epidermal cells throughout the leaf, 3) plasmodesmata between epidermal and ground tissue at the leaf margins, 4) intercellular spaces between epidermal and ground tissues, 5) plasmodesmata and intercellular spaces within the ground tissue, 6) plasmodesmata and sieve pores within the vascular tissue, 7) protoxylem lacunae with wax impregnated, thickened cell walls in the central vein, 8) suberin in the epidermis and bundle sheath cells, 9) starch-rich chloroplasts in parenchyma cells that line the sieve tubes of transverse veins, 10) four chloroplast morphologies that are tissue

74 specific and differences in the chloroplast grana of the adaxial and abaxial surfaces and 11) two modes of intercellular space formation including detachment of a discrete pectin plug and schizogony of intercellular walls.

The outermost layer of H. hawaiiana leaves is a thin amorphous cuticle. This structure is the most common cuticle type found in the seagrasses and contrasts with those found in some species that are porous, highly structured and/or with subcuticular cavities (Barnabas et ale 1977, Doohan and Newcomb 1976, Gessner 1971, Jagels

1973, 1983, Kuo 1978). Functionally the thin cuticle may impart selectivity to the leaf surface but must be permeable to water, gas and solutes (Arbor 1920, Mauseth 1988,

Sculthorpe 1961). This layer retards diffusion of some solutes into the cell, but ca. 25% of ions freely enter the

epidermis through the cuticle (Raven 1984). The underlying outer cell wall of H. hawaiiana is thickened, relative to the adjacent anticlinal and mesophyll

cell walls as commonly occurs in the seagrasses. The

thickened wall is thought to be the chief mechanical tissue

of the leaf blade that, along with turgor, maintains

rigidity of the leaf (Tomlinson 1980). Waldron et ale

(1990) has proposed the seagrass cell wall is a feature

adaptive to the marine environment due to its chemical

nature.

75 The thickest cell wall occurs at the absorbing surface and has been regarded as counteradaptive in terms of diffusion of solutes from the bulk medium to the plasmalemma (Raven 1984). The localization of suberin in the outer epidermal wall of H. hawaiiana may further illustrate an apparent aspect of counteradaptiveness in that the water repelling surface is also the absorbing surface for the seagrass. Alternatively, the seagrass epidermis may be a site of selection rather than a barrier. Pectic compounds, a reported component of the seagrass cell wall, function in cell wall porosity, the surface charge that modulates wall Ph and ion balance and perhaps recognition phenomena involved in symbiosis (Salisbury and Ross 1992). Suberin is a plant protection coating made of waxy lipid and phenolic compounds (Salisbury and Ross 1992). Suberin limits the diffusion of some solutes, but is freely permeable to gasses. Solereder (1913) described the epidermis of H. ovalis as "unusual" because of the occurrence of ring-like areas in the periclinal walls. Birch (1974) determined that a "ring like annulus" in the outer wall reacted strongly to silver nitrate and was so implicated in ion interactions. Tubules within the annulus were the reactive sites. The silver was thought to bind with chlorides in that they constitute 6 to 12 percent of the leaf dry matter and move easily to and from the leaves. The presumed function of the annulus in fl.

------ovalis is similar to one speculative function of the cell wall ingrowths found for putative transfer cells in H. hawaiiana, H. ovalis and other seagrasses. The chloride content of macrophytes in general affects cell turgor and the concentration changes as a function of external salinity (Lobban et ale 1985). Putative transfer cells illustrated here in the epidermal cells of H. hawaiiana, are a means by which plasmalemma area is increased (Tomlinson 1980). The addition of plasmalemma area is considered an overt morphological response by aquatic plants to resource availability (Raven 1984). According to Gunning and Pate (1969) " ••. in every instance where this distinctive morphology has been reported, a distinctive role in intensive transport of solutes can be postulated". Ingrowths develop as the leaf matures and requires intensive transport to support rapid leaf formation. These structures are best developed on the faces of the cell presumed to be most active in solute transport from the external environment. The form, frequency and final degree of development are characteristic of the location of these cells and the plant species (Gunning and Pate 1969). Seagrasses with putative transfer cells include Thalassia testudinum Banks ex Konig (Jagels, 1973, 1983), T. hemprichii (Ehrenb.) Aschers., Cymodocea serrulata (R. Br.)

Aschers., ~. rotundata Ehrenb and Hempr. ex Aschers. (Doohan

------and Newcomb 1976), Zostera campensis Setchell (Barnabas et al. 1977, 1980, 1982), z. marina L. (Jagels 1983), Z. muelleri Irmisch ex Aschers., Phyllospadix iwatensis Makino and £. japonicus Makino (Kuo et al. 1988). The physiology of species with or without this structure has not been compared.

Probable roles for putative transfer cells in seagrasses include: osmoregulation (Gunning and Pate 1969,

Jagels 1973, 1983, Birch 1974, Gunning 1977, Barnabas 1982,

Barnabas et al. 1982); a microenvironment for metabolic transformation of solutes (Pate and Gunning 1969, Raven

1984); absorption of solutes from the external environment

(Gunning and Pate 1969, Gunning 1977); nutrient exchange

(Gunning and Pate 1969, Gunning 1977); ion transport (Birch

1974, Faraday and Churchill 1979, Schroeder and Thorhaug

1980); gas exchange (Gunning and Pate 1969, Gunning 1977); cell secretion (Gunning and Pate 1969, Gunning 1977); assimilation (Tomlinson 1980); a site for enzymes (eg. phosphatase, Gunning et al. 1968) and polyions that may attract and concentration counter-ions close to the active membrane (Gunning et al. 1968); acquisition of inorganic carbon in photosynthesis (Gunning and Pate 1969, Jagels

1983) and the formation of acid and alkaline zones on the macrophyte surface that involves the use of bicarbonate

(Gunning and Pate 1969, Raven and Smith 1980, Jagels 1983,

78 Raven 1984) and polysaccharide accumulation (Pate et ale 1969). Environmental factors have been considered important in formation of transfer cells. Temperature and salinity are reported to influence cell wall elaboration in some systems (Jagels 1983). Wimmers and Turgeon (1991) demonstrated that aSYmmetrical distribution of cell wall elaborations in land plants relate to light. The elaborate development and distribution of putative transfer cell components in the upper leaf surface of H. hawaiiana may relate to greater metabolic activity associated with higher photon flux incident on those cells. This is yet to be tested. The distinct cell wall ingrowths with invaginated plasmalemma and closely associated organelles that characterize the adaxial epidermal cells of H. hawaiiana are suggestive of an intracellular flux system as described by Raven (1984). The abundance of mitochondria may be part of an ATP-requiring transmembrane movement of ions that may be occurring (Jagels 1973). The chloroplasts are said to be the energy source for photoaccumulation of ions in water plants (Gunning and Pate 1967). Amplification of plasmalemma area per unit cell surface area are frequently found in situations where large fluxes occur between symplast and apoplast (Gunning 1977). The structures of both of those systems are common in the leaves of R. hawaiiana.

79 The apoplast and symplast are two major components of plants (MUnch 1930) and constitute modes of intercellular channeling. The apoplast is composed of nonliving parts of plant body including the cell wall and intercellular spaces (external to the plasmalemma) while the symplast interconnects protoplasts with a plasmalemma continuum. The apoplast is further defined as the aqueous phase in the cell wall in which water content, pore size and fixed charge density are important (Raven 1984). The apoplast is involved in radial transport of nutrients. structures involved in the apoplast system include the cuticle, the thickened epidermal wall, the periplastic space formed between the cell wall ingrowths and the plasmalemma, the anticlinal walls of the epidermis, the mesophyll walls, intercellular spaces that are C0mmon between the epidermis and mesophyll as well as within the mesophyll, and the vacuolate mesophyll cells themselves. The innermost layer of the apoplastic pathway, the xylem, is restricted to the central vein in H. hawaiiana and other seagrasses. Symplastic intercellular connections are common within tissues and between tissues at the leaf margins. Branching plasmodesmata adjoin vascular parenchyma and sieve tubes, and callose lined pores in sieve plates are distinctive, especially in the transverse veins. The transverse veins appear to be specialized for translocation of sugars while

80 the central veins appear to be specialized for support and apoplastic flux. In general terms, the anatomy of H. hawaiiana as a marine seagrass can be contrasted with that of the algae and terrestrial vascular plants. As the seagrasses are aquatic plants with terrestrial origin (secondary hydrophytes), they retain the same basic systems as terrestrial plants. Although the cuticle of vascular plants is a barrier to water loss, the cuticle of seagrasses is thin and not a significant barrier to transport of nutrients and water. Algal cuticles are not chemically similar to that of the vascular plants (Lobban and Wynn 1981). stomata are specific sites of gas exchange in terrestrial vascular plants that are absent in the seagrasses and algae. Gas exchange occurs at the seagrass and algal surfaces; intercellular gas spaces are more significant in submerged than terrestrial plants, but for seagrasses they transport oxygen for root metabolism. The spaces function in buoyancy, radial and longitudinal transport of gasses. Gas spaces in algal cells are involved in buoyancy regulation and light scattering rather than in distribution of metabolic gasses as occur in analogous structures in vascular plants (Raven 1977). Intercellular transport of solutes and solvents occurs by plasmodesmata or microplasmodesmata in terrestrial and aquatic vascular plants and the algae. While the reduction in xylem in the

81 seagrasses shows that water is not transported upward as in terrestrial plant leaves, the well developed phloem ensures downward movement to the rhizome of photosynthate from the leaves.

concl.usions Plants with cell wall ingrowths and hence, protoplasts with high surface-to-volume ratios are found in most major taxa of multicellular plants (Gunning and Pate 1969). The occurrence of transfer cells in phylogenetically diverse assemblages of plants may represent a fundamental evolutionary solution to problems related to efficient and rapid short-distance transport of solutes (Gunning et al. 1968). Intracellular flux systems include the putative transfer cells and other organelles in near proximity to each other. The ultrastructure of H. hawaiiana commonly shows structures of symplastic and apoplastic flux systems. In addition to elaborately developed cell wall ingrowths, internal to the upper epidermal wall, the structures of apoplastic and symplastic systems are common within the leaves, well developed and distinctly distributed. The evidence for the function of symplastic intercellular translocation and/or communication is the presence of suberin in two tissues, the epidermis and the inner ground tissue of the bundle sheath cells. Evidence for apoplastic pathways is the abundant system of intercellular spaces and

82 abundant cell walls in the mesophyll that do not have barriers (plasmodesmata or water repelling chemistry) to non-protoplasmic flux. Both systems occur at the leaf margins, nearly within the same cell. The distinctive distribution of specialized cells within the leaf is indicative of specific areas of function within the leaf. Apparent water directing pathways occur at the epidermis, the bundle sheath cells and at sites of intercellular connections. The distinctive distribution of chloroplasts with lipid, or lipid + starch, or lipid + starch + protein may also infer special physiology related to each tissue. Specific questions raised by the results of this study could be pursued further by combining morphometric analysis of structures with probes into the physiology of the plant, in part and as a whole.

83 Figure 2.1. Habit of Halophila hawaiiana. Scale leaves (sL), foliage leaves (Fl), are adjoined to the rhizome (rh) at discrete nodes (no) by a petiole (pe). Roots (ro) also grow from the nodes. The space of the internode (In) coincides with the plastochrone interval. Scale = 7 cm. Figure 2.2. Leaf morphology and venation pattern. Transverse veins (Tv) branch from a central vein (CV) and are bounded by the marginal vein (Mv). Scale = 7 cm.

84 Cv

85 Figure 2.3. Transverse section oriented as upper and lower surfaces, line drawing. Upper and lower layers of epidermal cells (Ep) form a biseriate leaf except in areas of the central vein (Cv), transverse veins (Tv), and marginal veins (Mv). Aerenchyma (Ac) is concentrated in the central region. Parenchyma cells (Pa) line the large air canals and the mid vein. Scale = 20 ~m. Figure 2.4. Surface view of leaf cell pattern. Cell pattern determined by the pattern of anticlinal walls (arrow) shows isodiametric cells between veins and more elongate cell over veins. Scale = 20 ~m.

86 2.4

87 Figure 2.5. The epidermis I. A profile of adaxial epidermal cell periphery. A cuticle with small striations (arrow) overlays the cell wall composed of outer wall (ow) and inner wall (iw). Inner wall extends into the cytoplasm via ingrowths (I) that are lined with invaginated plasmalemma (arrowhead). Mitochondria (m) and chloroplasts (cl) cluster near the cell wall ingrowths. Scale = 0.8 ~m.

Figure 2.6. The epidermis II. A profile of abaxial epidermal cell periphery. striations or pores (arrow) in the cuticle give slight structure to the electron translucent cuticle. Cytoplasm contains nucleus (N), chloroplasts with highly organized grana (arrowhead), mitochondria, and cell wall ingrowths. Scale = 1.0 ~m.

88 i ' l 1

89 \0 o

Figure 2.7. The epidermis III. A network of fibrillar ingrowths and organelles adjacent to the inner epidermal cell wall. Ingrowths (I) are lined with an extracytoplasmic space (arrow) and closely associated with plasmalemma (double arrow), mitochondrion (m), chlo-roplast (cl), upper leaf surface. Inner wall (iw) fibrils form a "herringbone" patterns. Adaxial surface. Chloroplast in oblique face view. Scale = 0.16 ~m. Figure 2.8. The epidermis IV. Cytoplasm internal to inner epidermal cell wall; endoplasmic reticulum (er), ribosomes (r) and microtubules (arrow) internal to cell wall area without ingrowths. Adaxial surface. Scale = 0.13 ~m.

Figure 2.9. The epidermis V. Inner epidermal cell wall and cytoplasm, abaxial surface. Electron dense outer wall (ow) borders the inner wall (iw) which extends into cell wall ingrowths (I). Dictyosome derived vesicles are adjacent to the wall ingrowths (arrow). Scale = 0.25 ~m. 2.10

U> I-' Ep . ~ ., :.~~. -, ~~.< '~'.:\~~" ... .. -.'" .,'..' .',' .' "'l. ~j Figure 2.10. Symplastic connections within the epidermis I. Simple plasmodesmata cluster (arrow) in anticlinal wall between epidermal cells (Ep). Inner wall (iw) area distinct from middle wall (mw) that is a continuation of the outer epidermal wall. The middle wall coincides with the middle lamella. Scale = 0.8 ~m. Figure 2.11. Symplastic connections within the epidermis II. Plasmalemmas extend to adjacent protoplasts through plasmodesmata (arrow). Scale = 0.19 ~m. Figure 2.12. Symplastic connections within the epidermis III. An aggregate of plasmodesmata in the patterned inner surface of the anticlinal wall (iw) as seen in oblique face view. Scale = 0.19 ~m. ...,.... ~ ....• 2.13 2.14 ~\ . - ' .' - " . " : ' . " ". 10;:;::: . .... -'. ~.4" "'>w.. rr" •.-

1.0 N

Figure 2.13. Transverse vein ground tissue I. Parenchyma cell, two cells inward from the epidermal later, is adjoined to other parenchyma cells by both plasmodesmata (arrow) and intercellular spaces (As). Mitochondria (m) and chloroplasts (cl) are closely associated and bounded internally by the tonoplast (double arrow). The central vacuole (V) constitutes most of the cell volume. Scale = 1.3 ~m. Figure 2.14. Transverse vein ground tissue II. Plasmodesmata between highly vacuolate parenchyma cells in the ground tissue. Partial view of plasmalemma between adjacent cells (arrow). Scale = 0.19 ~m. Figure 2.15. Transverse vein ground tissue III. Endoplasmic reticulum (er) and plasmalemma (arrow) traversing wax impregnated wall of parenchyma cells near the vascular tissue. Scale = 0.19 ~m. Figure 2.16. Transverse vein vascular tissue I. Thin walled, large, vacuolate parenchyma cells (Pa) are external to undifferentiated cells of the bundle sheath (Bs). Chloroplasts with starch and protein inclusions (arrow) occur in this cell layer. The walls of sieve tube cells (st) are generally thicker and less electron dense than those of external cells. Proteinaceous fragments of sieve tube plastids (arrowhead) and empty sieve tube plastid envelopes (e) are scattered in the phloem. Scale = 4.0 ~m.

Figure 2.17. Transverse vein vascular tissue II. Additional structures from the same bundle (Fig. 16 further into the tissue) include: plasmodesmata (P), a sieve plate with a sieve pore (arrow) between sieve tube cells and a companion cell (lower left). Scale = 4.0 ~m.

Figure 2.18. Transverse vein vascular tissue III. Sieve pores (arrow) in the sieve plate between sieve tube cells are lined with fibrillar p protein-like material. Fragments of the sieve tube plastids (pr) and empty sieve tube plastid envelopes (e) are evident near the sieve pores. Plasmodesmata in face view (joined arrows) line the cell wall between phloem parenchyma (Pp) and a sieve tube cell. Scale = 2.0 ~m.

Figure 2.19. Transverse vein vascular tissue IV. Oblique section through a companion cell (Cc) and sieve tube cell with plasmodesmata in face and partial transverse views (joined arrows). Dense cytoplasm of the companion cell contains ER with ribosomes, vesicles (ve), larger vacuoles (V) and a microbody (rob). Scale = 0.3 ~m. Figure 2.20. Transverse vein vascular tissue V. A companion cell adjoins a sieve tube cell with by a branched plasmodesma. Callose (arrow) is formed on the sieve tube cell-side. The cytoplasm contains endoplasmic reticulum with ribosomes (r), vacuoles (V) and microbodies (rob). Scale = O. 3 ~m.

93 94 Figure 2.21. Transverse vein vascular tissue VI. Highly differentiated phloem from a transverse vein includes a companion cell (Fig. 20), sieve tube cells (st) and phloem parenchyma (Pp). Symplastic connections within phloem include a branched plasmodesma between a companion cell (Cc) and sieve tube cell, callose lined pores between sieve tube cells and simple plasmodesmata (arrow) between phloem parenchyma cells. Scale = 1.0 ~m. Figure 2.22. Transverse vein vascular tissue VII. Early stage of callose formation in a branched plasmodesma. Vesicles (ve) cluster at the location of deposition. Scale = 0.19 ~m.

95 2.21

96 \0 -..J

Figure 2.23. Marginal vein intercellular connections. symplastic connections (P) between epidermal cells of the leaf margin and the underlying large parenchyma cells that adjoin the vascular tissue. Scale = 0.6 ~m. Figure 2.24. Central vein vascular tissue I. Sieve tube cells (St) are internal to abundant parenchyma cells (Pa) with thin, electron dense walls and little cytoplasm. A sieve pore adjoins sieve tube cells. Undifferentiated bundle sheath cells (Bs) link to external parenchyma (Pa) by intercellular spaces (arrows). A protoxylem cell (X) is adjacent to 2 vascular parenchyma cells (Vp). Scale = 4.0 ~m. Figure 2.25. Central vein vascular tissue II. The developing sieve pore is lined with callose and the remnants of a desmotubule are evident (arrow). Scale = 0.25 ~m. Figure 2.26. Central vein apoplasm I. Large air canal (Ac) near the central vein vascular tissue is lined by similarly sized parenchyma cells (Pa) with little cytoplasm and unevenly thickened walls. A protoxylem cell (X) with thickened walls is adjacent to vascular parenchyma (Vp). Scale = 5.0 JLm. Figure 2.27. Central vein apoplasm II. The protoxylem cell wall contains wax (arrow) and electron dense material. Scale = 0.25 JLm. Figure 2'.28. Central vein apoplasm III. An electron dense, homogeneous "plug" occurs in the thickened cell walls lining large air canals (Ac). Scale = 0.5 JLm. Figure 2.29. Central vein apoplasm IV. Schizogonous formation of intercellular spaces near large air canals. Scale = 0.5 JLm.

98 Ac 2.27

2.28 2.29

-.~ Ac ;.

~. '..

99 Figure 2.30. Chloroplast morphology in abaxial epidermal cells: Epidermal cell chloroplast with lipid (arrow) in stroma between thylakoids (T). Nucleoplasm (n) is central in this view. The chloroplast is peripheral in the cell. Scale = 0.2 JLm. Figure 2.31. Chloroplast morphology in ground tissue I. Chloroplast with lipid (arrow), peripheral in parenchyma cell near epidermis. Scale = 0.8 JLm. Figure 2.32. Chloroplast morphology in ground tissue II. Parenchyma cells closely associated with a transverse vein. A large starch grain (8) occurs along with lipid (arrow). The chloroplast is central in the cell. Scale = 0.19 JLm. Figure 2.33. Chloroplast morphology in ground tissue III. A bundle sheath cell from a transverse vein contains three chloroplasts (1, 2, 3) that occupy most of the cell volume. Protein and starch-rich inclusions (arrow) as well as lipid occur. Scale = 1.6 JLm.

100

------2.30

101 Literature cited

Arbor, A. (1920). Water plants: a study of aquatic angiosperms. University Press, cambridge.

Barnabas, A. D. (1982). Fine structure of the leaf epidermis of Thalassodendron ciliatum (Forsk.) den Hartog. Aquat. Bot. 12:41-55. .

Barnabas, A. D. (1988). Apoplastic tracers studied in the leaves of a seagrass. I. Pathway through epidermal and mesophyll tissues. Aquat. Bot. 32:63-77.

Barnabas, A. D., Butler, V., steinke, T. D. (1977). Zostera capensis Setchell. I. Observations on the fine structure of the leaf epidermis. Z. Pflanzenphysiol. Bd. 85.S.:417 427.

Barnabas, A. D., Butler, V., steinke, T. D. (1980). II. Fine structure of the cavities in the wall of leaf blade epidermal cells. Z. Pflanzenphysiol. 99:95-103.

Barnabas, A. D., Butler, V., steinke, T. D. (1982). Zostera capensis Setchell. Some aspects of wall ingrowth development in leaf blade epidermal cells. Protoplasma 110:87-94.

Behnke, H.-D. (1981). Sieve element characters. Nor. J. Bot. 1:381-400.

Benedict, C. R., Scott, J. R. (1976). Photosynthetic carbon metabolism of a marine grass. Plant Physiol. 57:876-880.

Beveridge, T. J. (1989). The structure of bacteria. In: Poindexter, J. S., Leadbetter, E. R. (eds.) Bacteria in nature vol. 3; structure, physiology, and genetic adaptability. Plenum Press, New York.

Birch, W. R. (1974). The unusual epidermis of the marine angiosperm Halophila Thou. Flora Bd. 163:410-414.

Blanquet, P. R. (1976). Ultrahistochemical study on the ruthenium red surface staining. II Nature and affinity of the electron dense marker. Histochemistry 47:175.

Borowitzka, M. A., Lethbridge, R. C. (1989). Seagrass epiphytes. In: Larkum, A. W. D., McComb, A. J., Shepherd, s. A. (eds.) Biology of seagrasses; a treatise on the biology of seagrasses with special reference to the Australian region. Elsevier, Amsterdam.

102 BUlthuis, D. A. (1983). Effects of in situ light reduction on density and growth of the seagrass Heterozostera tasmanica (Martens ex Aschers.) den Hartog in Western Port, victoria, Australia. J. Exp. Mar. BioI. Ecol. 67: 91-103.

Cambridge, M.L., Kuo, J. (1982). Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae) III. Posidonia sinuosa Cambridge & Kuo. Aquat. Bot. 14:1-14.

Chafe, S. C. (1974). Cell wall structure in the xylem parenchyma of Cryptomeria. Protoplasma 81:63-76.

Chapman, A. R. O. (1979). Biology of seaweeds. University Park Press, Baltimore.

Colombo, P. M., Rascio, N., Cinelli, F. (1983). Posidonia oceanica (L.) Delile: A structural study of the photosynthetic apparatus. Mar. Ecol. 4:133-145.

Dahlgren, R. M. T., Clifford, H. T., Yeo, P. F. (1985). The families of Monocotyledons: structure, evolution and taxonomy. Springer-Verlag, Berlin. den Hartog, C. (1970). The seagrasses of the world. North Holland Publ., Amsterdam.

Doohan, M. E., Newcomb, E. H. (1976). Leaf ultrastructure and 13C values of three seagrasses from the Great Barrier Reef. Aust. J. Plant. Physiol. 3:9-23.

Doty, M. S., Stone, B. C. (1966). Two new species of Halophila (Hydrocharitaceae). Brittonia 18:303-306.

Esau, K. (1977). Anatomy of seed plants. 2nd ed. John Wiley & Sons, New York.

Faraday, W. E., Churchill, A. C. (1979). Uptake of cadmium by the eelgrass Zostera marina. Mar. BioI 53:293-298.

Gessner, F. (1968). Die Zellwand mariner Phanerogamen. Mar. BioI. 1:191-200.

Gessner, F. (1971). The water economy of the seagrass Thalassia testudinum. Mar. BioI. 10:258-260.

Gunning, B. E. S. (1977). Transfer cells and their roles in transport of solutes in plants. Sci. Prog. 64:539-568.

103

------Gunning, B. E. S., Pate, J. s., Briarty, L. G. (1968). specialized "Transfer Cells" in minor veins of leaves and their possible significance in phloem translocation. J. Cell BioI. 37:7-12.

Gunning, B. E. S., Pate, J. S. (1969). "Transfer Cells" plant cells with wall ingrowths, specialized in relation to short distance transport of solutes--their occurrence, structure and development. Protoplasma 68:107-133.

Haberlandt, G. (1914). Physiological plant anatomy. 4th Ed. (trans.). London: Macmillan and Co.

Jagels, R. (1973). studies of a marine grass, Thalassia testudinum I. Ultrastructure of the osmoregulatory leaf cells. Am. J. Bot. 60:1003-1009.

Jagels, R. (1983). Further evidence for osmoregulation in epidermal leaf cells of seagrasses. Am. J. Bot. 70:327 333.

Johansen, D. A. (1940). Plant microtechnique. McGraw-Hill, New York.

KUo, J. (1978). Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (posidoniaceae). I. Leaf blade and leaf sheath of Posidonia australis Hook F. Aquat. Bot. 5:171-190.

KUo, J., Aioi, K., Iizumi, H. (1988). Comparative leaf structure and its functional significance in Phyllospadix iwatensis Makino and Phyllospadix japonicus Makino (Zosteraceae). Aquat. Bot. 30:169-187. KUo, J. Ridge, R.W., Lewis, S. v. (1990). The leaf internal morphology and ultrastructure of Zostera muelleri Irmisch ex. Aschers. (Zosteraceae): a comparative study of the intertidal and subtidal forms. Aquat. Bot. 36:217 236.

Lobban, C. S., Harrison, P. J., Duncan, M. J. (1985). The physiological ecology of seaweeds. Cambridge University Press, London.

LUft, J. H. (1971). Ruthenium red and violet I: Chemistry, purification, methods of use for electron microscopy and mechanism of action. Anat. Rec. 171:327-368.

Mauseth, J. D. (1988). Plant anatomy. Benjamin/Cummings Pub. Co. Inc., Menlo Park.

104 McMillan, C., Williams, S. C. (1980). Sulphated phenolic compounds in seagrasses. Aquat. Bot. 8:267-278.

McMillan, C., Zapata, 0., Escobar, L. (1980). SUlphated phenolic compounds in seagrasses. Aquat. Bot. 8:267-278.

Munch, E. (1930). Die stoffbewegung in der pflanze. Jena, Gustav Fischer.

Naidoo, Y., Lawton, J. R., Barnabas, A. D., Coetzee, J. (1990). Ultrastructure and cytochemistry of squamulae intraviginales of the marine angiosperm, Halophila ovalis. S. Afr. J. Bot. 56:546-553.

Pate, J. S., Gunning, B. E. S. (1969). Vascular transfer cells in angiosperm leaves. A taxonomic and morphological survey. Protoplasma 68:135-156.

Pate, J. S., Gunning, B. E. S., Briarty, L. G. (1969). Ultrastructure and functioning of the transport system of the leguminous root nodule. Planta 85:11-34.

Posluszny, U., Tomlinson, P. B. (1991). Shoot organization in the seagrass Halophila (Hydrocharitaceae). Can. J. Bot. 69:1600-1615. Raven, J. A. (1984). MBL lectures in biology; energetics and transport in aquatic plants. Alan R. Liss. Inc., New York.

Raven, J. A., Smith, F. A. (1980). Significance of hydrogen ion transport in plant cells. Can. J. Bot. 52:1035-1048.

Reynolds, F. S. (1963). The use of lead citrate at high Ph as an electron opaque stain in electron microscopy. J. Cell. BioI. 17:208-212.

Sachet, M., Fosberg, F. R. (1973). Remarks on Halophila (Hydrocharitaceae). Taxon 22:439-443.

Salisbury, F. B., Ross, C. W. 1992. Plant physiology 4th ed. Wadsworth Pub. Co., Calif.

Schroeder, P. B., Thorhaug, A. 1980. Trace metal cycling in tropical-subtropical estuaries dominated by the seagrass Thalassia testudinum. Amer. J. Bot. 67:1075-1088.

Sculthorpe, C. D. (1961). The biology of aquatic vascular plants. Arnold Press, London.

105 Solreder, H. (1913). Systematisch-anatimische Infersuchung des blattes der Hydrocharitaceen. Beih. Int. Zbl 30:21 104. Tomlinson, P. B. (1972). On the morphology and anatomy of turtle grass, Thalassia testudinum (Hydrocharitaceae). IV. Leaf anatomy and development. Bull. Mar. Sci. 22:75-92. Tomlinson, P. B. (1974). Vegetative morphology and meristem dependence-the functional aspects of productivity in seagrasses. Aquaculture 4:107-130. Tomlinson, P. B. (1980). Leaf morphology and anatomy in seagrasses. In: Phillips, R. C., McRoy, C. P., (eds.). A handbook of seagrass biology. Garland STPM Press, New York. Tomlinson, P. B. (1982). Detailed descriptions of families, family hydrocharitaceae. In: Metcalfe, C. R. (ed.) Anatomy of the monocotyledons: The helobiae (Alismatidae) (including the seagrasses). Vol. VII. Clarendon Press, Oxford. Waldron, K. W., Baydoun, E. A. -H., Brett, C. T. (1989). Comparison of cell wall composition of tissues from the seagrasses Halophila and Halodule. Aquat. Bot. 35:209 218. Wimmers, L. E., Turgeon, R. (1991). Transfer cells and solute uptake in minor veins of Pisum sativum leaves. Planta 186:2-12.

106 CHAPTER 3

ULTRASTRUCTURE OF SEAGRASS AND EPIPHYTE INTERFACES FROM WAVE EXPOSED AND SHELTERED SUBTIDAL HABITATS

Abstract Electron microscopy observations demonstrated that means of attachment by a variety of epiphytes on leaves of Halophila hawaiiana Doty and stone differ among organisms and correspond to variable ultrastructural changes within the host epidermal cells. Additionally, patterns in distribution of colonists and ultimate epiphytes on same aged leaves varied between wave exposed and wave sheltered sites. Crustose coralline red algae and diatoms were primary colonists on seagrass leaves at the wave exposed site. Bacteria and diatoms were primary colonists on seagrass leaves at the wave sheltered site. Seagrass epidermal cells were modified by the presence of crustose coralline red algae as well as solitary and tunneling bacteria. Ultrastructural examination of leaves revealed that epidermal cells adjacent to crustose coralline red algae showed distinctive elaborations of the cell wall nearest the region of epiphyte attachment. Secretory organelles were numerous in these epidermal cells. In contrast, epidermal cells adjacent to bacteria showed disruption of the fibrillar cell wall in that region and modified elaborations of the cell wall nearest the region of attachment. Osmiophilic droplets and vesiculate membrane-

107 bound structures were also abundant in epidermal cells that were colonized by bacteria. Colonists common to leaf surfaces at both sites were diatoms, filamentous red algae, cyanobacteria and bacterial microcolonies. Ultrastructural examination of epidermal cells in regions of these organisms demonstrated that these epidermal cells were similar to those of unepiphytized leaves. Differences in the propagule pools and conditions for cell attachment in the contrasting environments resulted in differing epiphyte communities.

Introduction Theories of marine benthic community organization stress the importance of competition and predation, environmental gradients, disturbance, and early settlement events. Settlement rates and patterns have been shown, in some locations, to depend on propagule supply and substrate suitability (Connell 1985). Variation in the nature of the propagule and the rate of recruitment can contribute to species diversity, community structure and dynamics (Keough and Downes 1982). Roughgarden et ale (1986) developed a theory of "supply side ecology" that stresses the relationship between potential recruit abundance, resource limitation and competition in some communities. Taxonomic composition of the gene pool, hydrodynamics, physical characteristics of the SUbstrate, and disturbance influence recruitment and colonization (Palmer 1988). The success of

108 settlement, germination and recruitment processes by propagules determines algal establishment and may influence species distribution (e.g. McDermid 1988). The early inter-species associations in epiphyte community development are the focus of this study. The spatial relationships among these species are regarded as potentially important determinants of community structure and dynamics. The epiphytes, the cellular dynamics of the host and colonist interface and the environment are discrete factors that play major roles in setting the nature of each community. The importance of these three factors are elaborated in the following paragraphs.

Epiphytes Epiphytes are ubiquitous in seagrass ecosystems (Borowitzka and Lethbridge 1989). They contribute significantly to the total energy budget and species diversity of near-shore ecosystems (Murray and Wetzel 1987). This community has been defined as epiphytic algae plus any associated heterotrophic organisms, detrital material (Fountaine and Nigh 1983) and the host. Common autotrophic epiphytes include diatoms, cyanobacteria, macroalgae and encrusting algae. Common heterotrophic organisms include bacteria, fungi and invertebrates. Each epiphyte may be further epiphytized (Novak 1984) generating layers of organisms (Hudon and Bourget 1981). Thus the structure of

109

---- ... _-_. ------an epiphyte community on a substrate results from a complex series of processes and interactions (Hudon and Bourget 1981, D'Antonio 1985). The rapid growth rate and turnover of these organisms lend themselves to studies of community dynamics. In general, seagrass production may stimulate epiphyte production; transfer of limiting nutrients from seagrasses to epiphytes and/or the converse has been demonstrated in many systems (McRoy and Barsdate 1970, Harlin 1971, 1973, 1975, McRoy et ale 1972, McRoy 1974, McRoy and Goering 1974, Brylinsky 1977, Penhale 1977, Penhale and Smith 1977, Libes and Boudouresque 1987, smith and Penhale 1980). Dissolved organic matter released from the seagrass leaves into the surrounding water is reduced in the presence of epiphytic bacteria and algae (McRoy and Goering 1974). Transfer of carbon, nitrogen and phosphorus at the leaf surface is likely to enhance growth and primary production by epiphytes (Libes and Boudouresque 1987). Various patterns in the distribution and diversity of macro- and micro-algal epiphytes (Heijs 1985a,b, Ballantine 1979) as quantitative and qualitative epiphyte community components (Heijs 1985b) have been described; crustose coralline red algae are often spatial dominants while many species of filamentous red algae are common. Microbial populations on seagrass leaves vary in abundance and morphology with position on the host that coincide with

110 specific aged areas (Novak 1984). Factors underlying distribution patterns of epiphytes on seagrasses include host morphology, development, age (Borowitzka and Lethbridge 1989), and leaf longevity (McComb et ale 1981). Phenolic compounds (Harrison and Chan 1980, Harrison 1982, Mazzella et ale 1981) and "algal antibiotics" are released by some seagrasses (Burkholder 1973) and may markedly influence the diversity and composition of epiphytes on seagrass leaves (Sand-Jensen 1977, Ducker and Knox 1984). The close relationship of microbes with a seagrass host can involve exchange of metabolites and nutrients (Harlin 1971, 1973, McRoy and Goering 1974). Leaf exudates generally correlate with increased bacterial growth (Hough and Wetzel 1975, Kirchman et ale 1984, Moriarty and Iverson 1985). Intrusive anchorage structures between the host and epiphyte are not required for the exchange of nutrients/metabolites at the leaf surface (Harlin 1971, 1973). Although the general ecological significance of various microalgal epiphytes is still poorly documented (Murray and Wetzel 1987) the potential for interaction between a host and epiphyte, as well as between epiphytes, is great because of their spatial proximity (e.g. Wetzel 1983).

The host and epiphyte interface The interface between a host and epiphyte is defined here as those structures that are in physical contact as a

111

------result of cell to cell or extracellular attachments. How cell surface structures and extracellular polymeric secretions (EPS) from microbes regulate fine-scale processes of colonization by those organisms is well described for certain systems (e.g. Marshall et ale 1971a,b, Costerton and Irvin 1981, Mirelman and Ofek 1986, Decho 1990). Microbial attachment structures include fimbriae, antigens (Rosenberg and Kjelleberg 1986), lipopolysaccharides and extracellular polysaccharide glycocalyces (Costerton et ale 1985). Bacterial gylcocalyces are a nearly universal surface component of bacterial cells in the marine environment. On inert and living substrata, the extracellular polysaccharides bind cells together to form microcolonies (Costeron et ale 1981). On living substrata the attachment is reinforced by cell-surface polysaccharides (Fletcher and Floodgate 1973, Corpe 1980) and may involve specific glycoproteins (Fletcher and Marshall 1982). While the specific surface chemistry has not been determined for H. hawaiiana, some aspects of cell wall composition have been determined for other Halophila species. The cell walls of select species of Halophila contain cellulose, non-cellulosic sugars, uronic acids and pectin (Waldron et ale 1989). Suberin and pectin in the epidermal cell walls of H. hawaiiana, reported in Chapter 2, may be important in host-epiphyte and epiphyte interactions. Suberin contains phenolic compounds as part of its structure

112 and may be induced by a stimulus (Salisbury and Ross 1992). Pectic compounds affect wall porosity as well as surface charge and may, speculatively, function as recognition molecules to signal a developmental response to other cells (Salisbury and Ross 1992). A metabolically active interface is established between microbes and other colonists of seagrass leaves (Maki et ale 1990). Community development on a surface with microbes may be mediated by the microbial consortia as they metabolically interact and as their extracellular polysaccharides influence further microbial binding to the surface. Seagrass leaves provide unusually uniform surfaces for epiphytic growth in marine environments (Feldmann 1937, Penhale 1977, Kenworthy et ale 1989, Duarte 1989). The most metabolically active tissue in seagrass leaves is the outer layer of cells, the principal surface for colonization. This tissue is the major site for photosynthesis and respiration, nutrient exchange, cell secretion, absorption (Gunning and Pate 1969, Gunning 1977) and osmoregulation (Jagels 1973, 1983). These cells characteristically have cell wall ingrowths lined by plasmalemma. The term "transfer cell" has been applied to this morphology in other systems (Gunning and Pate 1969), but the function of the cell wall ingrowths in seagrass cells has generally not been determined. Harlin (1971) proposed that the passage of

113 nutrients and photosynthates to epiphytic algae may involve this morphology.

Environmental influences on settlement of epiphytes Environmental factors can have marked influence on spatial and temporal patterns of epiphyte distribution on on seagrass leaves (Fountaine and Nigh 1983, Borowitzka and Lethbridge 1989). Fine-scale environmental variations create microhabitats for ranges in temperature, salinity, irradiance, and nutrient input; these microhabitats may be sufficient to explain patchiness in settlement and ultimate epiphyte growth. In general, surfaces in aquatic environments are attractive to bacteria because of the higher concentrations of inorganic and organic nutrients at these interfaces (Costerton et ale 1981). Water passing over the substratum is the principal mechanism whereby micro-organisms and nutrients come into contact with substrata. Water flow regimes can influence the thickness of the boundary layer, the time a propagule resides at a surface, and consequently a propagule's ability to adhere, attach and settle on a surface (see Lobban et ale 1985). The taxonomic composition of the propagule pool, physical characteristics of the substrate and hydrodynamics of the field site are themselves dynamic and complicate understanding of colonization in benthic marine environments (Palmer 1988).

114 The major objective of this study is to assess the colonization process of similar substrates, mature seagrass leaves, in proximal sites along an 800 meter environmental gradient from wave exposed to sheltered sites. This investigation is organized around the null hypothesis: wave exposure does not influence colonization of seagrass leaves by epiphytes. Additionally, the sequential null hypotheses are considered: 1) there are no differences in the effects of colonists on host tissues at these two sites, 2) there are no differences in the specific affects of different taxa of colonists on the host, and 3) there are no differences in subsequent colonization depending on taxa of colonists present. Varied theories of community ecology consider the importance of settlement interactions as driving forces behind community development and structure (see Chapter 1). By assessing the fine structure of potentially interacting cells in this host and epiphyte interface, I describe for the first time the nature of several cell to cell interactions during early stages in the establishment of the seagrass and epiphyte community development.

115 Materials and Methods sampling and site Epiphytized leaves of Halophila hawaiiana were collected from the same populations of wave exposed (Malaekahana) and wave sheltered (Moku'auia) sites during the same time as samples for fine structural observations (Chapter 2). Specific growth rate was determined by tagging plant apices as described in Chapter 2. The leaf density was determined by harvesting and counting leaves from 5 quadrats (500 cm2) for each site. The number of leaves per m2was counted and the mean and standard deviation were calculated (Table 3.2). The percent cover of micro epiphytes was estimated on 100 leaves from each of the 5 quadrats/site with the use of a Bausch and Lomb dissecting microscope, at 40X total magnification. Leaves were considered epiphytized if at least 25% of the surface were colonized. Leaves were further observed at the cellular level via electron microscopy. Seawater temperatures and salinities during this study were as reported in Chapter 2.

Sample preparation for electron microscopy Pre-senescent but mature, epiphytized leaves ca. 12 days age were fixed for transmission and scanning electron microscopy (TEM and SEM). Five fixations were designed to stabilize extracellular microbial matrices as well as the prokaryotic and eUkaryotic cells. Results of those

116 fixations are reported for TEM. The primary and secondary fixations were fundamentally similar to those described in Chapter 2. Prefixation components varied as follows:

Fixation~. Standard fixation. The fixation components were 2% glutaraldehyde buffered in 0.1 M sodium cacodylate to pH 7.6 in filtered seawater.

Fixation~. Polysaccharide stabilization including bacterial capsules, extracellular mucopolysaccharides and acidic polysaccharide (Jones et ale 1969, Fletcher and Floodgate 1973, Khan et ale 1990). Ruthenium red was added at a concentration of 1% to 2% glutaraldehyde buffered to pH 7.6 with 0.1 M sodium cacodylate in filtered seawater following Luft (1971) and Blanquet (1976).

Fixation~. Microbial capsule stabilization. Lysine was added at a concentration of 50 mM to 2% glutaraldehyde buffered to pH 7.6 with 0.1 M sodium cacodylate in filtered seawater following Jacques and Graham (1989), Akin and Rigsby (1990). Fixation Q. Complex carbohydrate stabilization and localization. A 2% solution of tannic acid was added at a concentration of 2% glutaraldehyde buffered to pH 7.6 with 0.1 M sodium cacodylate in filtered seawater following Sannes et ale (1978), McCarthy et ale (1979), Singley and Solursh (1980) and Khan et ale (1990).

Fixation~. Polysaccharide stabilization and complex carbohydrate localization. Ruthenium red and tannic acid

117

~----_._--- -- were added at concentrations of 1% and 2% respectively to 2% glutaraldehyde buffered to pH 7.6 with 0.1 M sodium cacodylate in filtered seawater. Leaves or leaf sections were prepared for SEM and TEM by Fixation A, critical point dried via use of an Autosamdri critical Point Dryer, model-810, and sputter coated in a

Hummer sputter Coater, model II with ca. 30 nm gold palladium and viewed in a Hitachi 5-800 Field Emission Scanning Electron Microscope.

Epiphyte population counts The epiphyte load was determined by placing epiphytes into taxonomic categories and then measuring their cell density per host cell per site from micrographs or direct observation by TEM (Table 3.2). The categories were: 1) crustose coralline red algae (Rhodophyta), 2) filamentous red algae (Rhodophyta), 3) diatoms (Bacillariophyta), 4) blue-green bacteria (Cyanobacteria), 5) solitary bacteria attached to the leaf surface, 6) solitary bacteria associated with the leaf surface but not attached directly,

7) bacterial cells in microcolonies, attached directly to the leaf surface, 8) bacterial cells in microcolonies associated with the leaf surface but not attached directly and 9) invasive bacteria or those present in regions of the cell wall and/or cytoplasm. Spatial relationships and means of attachment were noted for each category.

118

._------To assess patterns between epiphytes and seagrass leaf anatomy for each site, wall ingrowth layers, and chloroplasts were counted per host cell for 100 cells. Those variables were counted from unepiphytized (less than 5% cover) and epiphytized (ca. 25% cover) leaves (Table 3.3). Each category was analyzed as a set of four including 1) epiphytized leaves from Malaekahana, 2) unepiphytized leaves from Malaekahana, 3) epiphytized leaves from Moku'auia, 4) unepiphytized leaves from Moku'auia. statistically significant differences were evaluated by a Kruskal-Waller nonparametric analysis of variance. Groups that showed significant differences were evaluated further to determine significant differences among means by an g posteriori non-parametric equivalent to Fisher's Least Significance Difference Test (Dowdy and Weardon 1991) (Table 3.4).

119 Results Fixation results, leaf densities per site, and percent cover of epiphytes on leaves for both sites are presented in Part A. Epiphytes, their spatial relationships on H. hawaiiana leaves and evidence for modifications of the host epidermis with epiphytism by specific taxa are presented in Part B.

Part A. Technical and quantitative evaluation of epiphytes and the host

Fixation results Results from five fixations were compiled according to fixation quality of host and epiphyte cells, glycocalyces and cell products (Table 3.1). Information is not available for each taxon and each fixation because of the heterogeneous nature of each epiphyte community (Table 3.1). Fixation A produced satisfactory results for all the tissues examined. The seagrass cell wall was fibrillar (Figs. 3.7, 3.9, 3.11) and the inner wall ingrowths were electron dense (Figs. 3.3, 3.4, 3.9). Secretory organelles and products were evident (Fig. 3.10, 3.11). Diatom protoplasm and organic coating are shown in Fig. 3.5. Crustose coralline red algae and filamentous red algae (Fig. 3.1) were adequately preserved for the objective of this study. None of the fixations were designed for best

presentation of the caco3 impregnated cell wall. Bacteria 120 were generally well preserved, some fibrous glycocalyces including that of a cyanobacterium, were evident (Figs 3.19, 3.24). Ruthenium red (Fixation B) is thought to enhance the electron density of acidic polysaccharides. The fibrillar nature of the cell wall (Figs. 3.12, 3.30) was pronounced, and the cell wall invaginations were electron dense (Fig. 3.6, 3.12). Nuclei, chloroplasts with a dense stroma, other organelles and apparent secretory products also showed distinct electron density (Fig. 3.14). The fibrillar nature of microbe glycocalyces and the heterogeneous components of the bacterial capsules was best preserved by this fixation when compared with all other fixations (Figs. 3.17, 3.18, 3.21, 3.36). The results of Fixation C showed the fibrillar nature of the cell wall (Figs. 3.26, 3.27, 3.29, 3.31) and the cell wall ingrowths (Fig. 3.31). Crustose coralline algae (Figs. 3.26, 3.27, 3.29) were adequately preserved. Diatoms (Figs.

3.29-3.32) with secretion products (Figs. 3.32, 3.33) w~re presented clearly. Bacteria, and their inclusion bodies (Figs. 3.16, 3.35, 3.38), electron dense attachment mucilage and fibrillar glycocalyces (Figs. 3.26, 3.28, 3.29) as well as apparent metabolic products of tunneling bacteria (Figs. 3.26, 3.29, 3.33) were well preserved. The results of Fixation D were most distinctive when compared with other fixations. While overall presentation

121 of the seagrass was poor because of marked electron density of the cell walls (Figs. 3.25, 3.39, 3.42, 3.43), a benefit of this fixation is enhanced resolution of the complex polysaccharides in the cell wall matrix. The nucleus/nucleolus was well preserved but chloroplasts and mitochondria were poorly preserved. Apparent cytological modification induced by the bacteria contributed to the altered appearance of these epidermal cells (Fig. 3.25). Bacterial cells and capsules were well preserved by this method (Figs. 3.25, 3.37, 3.39-3.44, 3.46, 3.49). Means of attachment by bacteria to leaves (Figs. 3.25, 3.39-3.44) and heterogeneous glycocalyces (Figs. 3.37, 3.42, 3.44, 3.46, 3.47) were evident by this method of fixation. The results of Fixation E, as well as Fixation D, showed possible heteropolymer composition of bacterial glycocalyces and attachment mucilages for bacteria. These bacterial mucilages were electron dense in contrast with fibrous capsule components in an unattached region of the cell (Figs. 3.17, 3.18). Heterogeneous glycocalyces (Figs. 3.21, 3.22) were evident by this method of fixation. Bacterial cells were generally well preserved by this method (Fig. 3.17, 3.18,3.20,3.21,3.23,3.48).

~ Contrast of Seagrass and Epiphyte Populations/Site The mean density of leaves in the seagrass population at the wave exposed site (Malaekahana) was 985 leaves/m2 ± 233 S.D., n = 5 quadrats. The mean percent cover of

122 epiphytes on ca. 12 day old leaves was 23% ± 16 S.D., n = 5 quadrats. The mean density of leaves in the seagrass population at the wave sheltered site (Moku'auia) was 2920 leaves/m2 ± 411 S.D., n = 5 quadrats. The mean percent cover of epiphytes on ca. 12 day old leaves was 6% + 5.7 S.D., n = 5 quadrats. Leaf density and percent cover of epiphytes at each site were significantly different according to results from t-tests with C.I. greater than 99% for both leaf density and percent cover of epiphytes. The abundance of taxa within epiphyte communities was assessed as the number epiphyte cells in each taxonomic category per seagrass epidermal cell for 100 cells (Table 3.2). Epiphytes were counted on unepiphytized leaves (less than or equal to 5% cover) and epiphytized leaves (ca. 25% cover). As a preliminary assessment of site specific patterns in two cellular components, the number of 1) chloroplasts and 2) wall ingrowth layers per leaf cell was tabulated for unepiphytized and epiphytized leaves from each site (Table 3.3) • No statistical difference was found within the communities on epiphytized and non-epiphytized leaves from both sites for the following organisms: filamentous red algae, diatoms, cyanobacteria and microcolonies of bacteria attached to leaves (Table 3.4). Statistically significant differences were found within the epiphytized and

123 unepiphytized leaves for crustose coralline red algae, bacteria attached to the leaves, bacteria present but not attached to leaves, microbial colonies present but not attached to leaves, and invasive bacteria. Further analysis was carried out to determine which colonist or seagrass leaf response differed between non-epiphytized leaves and epiphytized leaves for each wave exposed and sheltered sites (Table 3.4). Distinctive trends for the wave exposed site include both the highest number of crustose coralline red algal cells and wall ingrowth layers per seagrass epidermal cell. The number of chloroplasts per seagrass epidermal cell were similar with slightly greater abundance in unepiphytized leaves. Distinctive trends for the wave sheltered site include the highest number of solitary and invasive bacteria per seagrass epidermal cell. The number of chloroplasts per seagrass epidermal cell differed between epiphytized and nonepiphytized leaves; unepiphytized leaves from this site showed the most abundant chloroplasts while the leaves epiphytized by bacteria showed the fewest chloroplast per host cell. No site specific difference occurred in solitary bacteria and bacterial in microcolonies associated but not attached to the leaves.

124

- .--_._.. _._------Part B. Qualitative observation on epiphytized seagrass leaves. The epiphytes The following information is primarily from transverse sections through the seagrass host and epiphyte interfaces; the community composition and spatial relationships among epiphytes and between the epiphytes and hosts are illustrated for leaves from wave exposed and sheltered sites. The epiphytes common to both sites are listed in the following figures: crustose coralline red algae (Figs. 3.1 3.5, 3.11, 3.26, 3.27, 3.29), filamentous red algae (Fig. 3.2), diatoms (Figs. 3.5, 3.6, 3.8, 3.26-3.33), cyanobacteria (Figs. 3.5, 3.18, 3.27, 3.29, 3.46), and other bacteria (Figs. 3.1, 3.2, 3.5, 3.13, 3.14, 3.19-3.50). Multilayered epiphyte communities on seagrass leaves from the wave exposed site were common (Figs. 3.1-3.5. Crustose coralline red algae were primary colonists and adnate to the leaf surface. Up to 4 cell layers were observed on the adaxial seagrass surface and 1 to 2 layers on the abaxial surface. Filamentous red algae (Fig. 3.2) diatoms (Fig. 3.5) and cyanobacterial cells within a fibrous polysaccharide matrix (Fig. 3.1) overlaid the crustose coralline algal cells. Prostrate cells of a filamentous red alga were anchored to the underlying algal cells at their cell walls (Fig. 3.2). The outermost epiphyte layer was generally bacteria (Figs. 3.1, 3.2). No intrusive

125 structures extended within the layers of colonists or between epiphytes and the host (Figs. 3.1-3.4). Bacteria dominated epiphyte communities on seagrass leaves from the wave sheltered site were common (Figs. 3.25

3.29). Bacteria were primary colonists attached to the seagrass surface and also found within diatom frustules

(Figs. 3.26-3.29, 3.31) as well as epidermal cell walls

(Figs. 3.26-3.33). Crustose coralline red algae were present as secondary or tertiary colonists at this site in contrast to being primary colonists at the wave exposed site

(compare Figs. 3.1-3.5 and Figs. 3.26-3.29). Microbes of varied morphologies and glycocalyces configurations occurred at both sites. Cyanobacterial cells of similar morphology and matrix were closely associated with crustose coralline red algal epiphytes (Figs. 3.1, 3.5,

3.27, 3.29) or attached to the leaf (Fig. 3.18). A larger multicellular cyanobacterium was associated with the abaxial seagrass surface (Fig. 3.46). The glycocalyces were 1) fibrillar and loosely associated with the leaf surface

(Figs. 3.1, 3.5, 3.29), 2) condensed outside a cleared or void area around the cells, 3) compacted around the cell wall (Fig. 3.46) as well as 4) heterogeneous at the attachment point (Fig. 3.18). Solitary bacteria with concentric, internal double membranes (Figs. 3.17, 3.21,

3.36 and 3.48) occurred at both sites. Bacteria of this morphology have been described as nitrifying bacteria

126 (Beveridge 1989). A notable polarity of the glycocalyx showed neutral mucopolysaccharides attaching the cell to the substrate (Figs. 3.17, 3.21) and in a similar unattached cell (Fig. 3.48). Other solitary bacteria with diverse cell and glycocalyx morphologies occurred attached and associated with the seagrass surfaces from both sites (Figs. 3.2, 3.5, 3.13, 3.14, 3.15-3.17, 3.19-3.22, 3.25-3.30, 3.34-3.37, 3.39-3.44, 3.46-3.49). A low magnification montage (Fig. 3.25) portrays a seagrass surface from the wave sheltered site congested with bacteria and mucilages. Distinct dividing cocci are shown in Figures 3.41-3.43.

The host and epiphyte interface The transmission electron micrographs in Chapter 2 illustrate the ultrastructure of unepiphytized leaves of H. hawaiiana from wave exposed and sheltered sites. The epidermal cells of epiphytized host leaves are contrasted with of unepiphytized host leaves here to examine cell surface and cytological modifications that coincided with the presence of specific colonists. The specific diatoms observed in this study did not alter the leaf's organelle abundance or distribution of the ingrowth region of the inner epidermal wall (Figs 3.6, 3.7, 3.13, 3.14). The fibrillar cell wall, abundant mitochondria and lipid-rich chloroplasts occurred in cells with diatoms

127 as primary colonists as in unepiphytized leaves (Figs. 3.6 with inset). The cuticle was generally intact; the small cuticle striations or possibly pores in the cuticle were occasionally more pronounced in areas underlying diatoms (Figs. 3.6, wave exposed and 3.30 wave sheltered) than in unepiphytized leaves (Chapter 2). Peg-like attachment structures permeated the cuticle and adjoined the frustule to the leaf (Fig. 3.8). Diatoms as primary colonists on the seagrass without associated bacteria were only observed on leaves from the wave exposed site. Modifications observed in the host cells with bacteria as well as diatoms (Figs. 3.26-3.33) were attributed to bacteria. Crustose coralline red algae were common primary colonists on leaves from the wave exposed site and secondary or tertiary colonists on leaves from the wave sheltered site. A leaf's organelle abundance and distribution of the ingrowth region of the inner epidermal wall were consistently modified when these algae were primary colonists on leaves (Figs. 3.3, 3.4, 3.10-3.12). The most conspicuous modification was the increased elaborateness of the ingrowth region of the outer epidermal wall. The number of "layers" of wall elaboration were significantly higher (Figs. 3.3, 3.4, 3.11) and the increase in abundance of organelles specialized for cell wall synthesis and secretion was marked (Figs. 3.10-3.12) when compared with unepiphytized leaves (Fig. 3.9). These organelles were also

128 abundant in areas of colonized cells between the elaborate ingrowths (Fig. 3.10, 3.11). Mitochondria (Figs. 3.3, 3.10, 3.12) and lipid-rich chloroplasts (Figs. 3.3, 3.10) were clearly present in the host cells. The cuticle is generally intact and the pits can be more pronounced at the leaf and algal interface (Figs. 3.4, 3.5, 3.10, 3.12). Diverse bacterial cell morphologies and glycocalyx morphs have been reported under the subheading "epiphytes". Host cell modifications were not observed when bacteria were in the phyllosphere, associated with the leaf rather than attached. Thus, bacteria in microcolony configurations and solitary bacteria as secondary or tertiary colonists did not correlate with persistent modification of seagrass cells in these mature leaves. A clear example of host cell modification is shown on a sparsely epiphytized leaf from the wave exposed site (Figs. 3.13,3.14). The distended ingrowth area of the inner epidermal wall occurred only where bacteria were embedded in the seagrass wall. The ingrowth region appeared lined with secreted material, perhaps tannin. Some lysis of the cell wall was apparent around the bacteria. The whole cell appeared unaltered except in regions adjacent to bacteria. The ingrowth region did not appear modified by the solitary attached bacteria or the diatoms. Bacteria within the outer epidermal walls were commonly observed in leaves from the wave sheltered site (Figs. 3.26-

129 3.33, 3.38). In each case the electron dense pectic matrix and cellulose fibrils were disrupted. The cell wall alteration appeared to progress from the outer wall to the inner wall as do the bacteria (Fig. 3.26). Bacteria appeared to erupt from the cell wall in an advanced stage of wall invasion. The cuticle was generally intact when bacteria were within the cell wall; striations/pores were as described for unepiphytized cells. The abundance of the pores increased near sites of cell wall lysis (Fig. 3.30).

The ingrowth region of the epidermis was modified when bacteria were in the epidermal cell wall. Abundant electron dense spheres were commonly adjacent to the ingrowths (Figs.

3.26, 3.31). Chloroplasts and mitochondria were less consistently in the cell periphery and secretory organelles were not evident.

The anatomy of seagrass leaves the wave sheltered site that were epiphytized by bacteria appeared very different from leaves epiphytized by crustose coralline red algae or invasive bacteria. Owing to the complexity of the bacterial community and the tannic acid fixation, the source and target of cytological modifications are difficult to assign

(Fig. 3.25). Distinctive modifications included an eroded cuticle, rare cell wall ingrowths but invaginated plasmalemma, no mitochondria, chloroplasts altered or not well preserved, electron opaque spheres (as occur with invasive bacteria), unusual membrane bound inclusions and

130 vesicles and large electron dense bodies (perhaps tannin). A closer look at the attachment of various solitary bacteria from this site shows the semi-embedded nature of the solitary bacteria at the host interface. The bacteria that appeared to cause the greatest host modification were surrounded by neutral and/or complex polysaccharides as determined by the positive responce to tannic acid within the fixation (Figs. 3.39-3.44). Those modifications included loss of the cuticle, as well as the above mentioned alterations (Fig. 3.25). Solitary bacteria adnate to leaves did not always coincide with modification of the interface, eg sparse bacteria attached to the leaf from the wave exposed site (Fig. 3.1, 3.18) although the neutral polysaccharide attachment of a bacterium from that site coincided with loss of the cuticle as occurred in the wave sheltered site (Figs. 3.17 wave exposed, 3.39-3.44 wave sheltered). Bacterial glycocalyces have been focused on in these results as a property of the microbial cell. Glycocalyx formation, as extracellular polymeric secretions (EPS) is also a property of the microenvironment of the cell and as such is listed here. Glycocalyces that stain positively with tannic acid were a common property of bacteria from the wave sheltered site (Figs. 3.25, 3.37, 3.39-3.44, 3.46 3.49). When present, the specific mucopolysaccharides often coincided with intercellular aggregations (Figs. 3.37, 3.47)

131 and epiphyte attachment to the host (Figs. 3.17, 3.18). Glycocalyces that stained positively with Ruthenium red formed part of heterogeneous mucopolysaccharides from the wave exposed site (Figs. 3.17, 3.18, 3.21) and were a common component of bacteria associated with the leaves in the phyllosphere (Figs. 3.20-3.22). A microcolony matrix and an outer fibrillar glycocalyx of a bacterium associated with the seagrass leaf from the wave exposed site are other examples mucopolysaccharides that stain positively with Ruthenium red (Figs. 3.48 and 3.50). A range of electron densities occurred in diverse bactrial glycocalyces without special probes for acidic or neutral polysaccharides (Figs. 3.2, 3.16, fixations A and C) which may identify proteinacous components.

Discussion This is a first report of anatomical patterns between a host and its epiphytes that range from no impact on the host interfacial cells, to consistent and distinctive protoplasmic alterations that result from the interspecific interactions. Four major patterns observed in this study include: 1) the primary layer of colonists on seagrass leaves differed between leaves from wave exposed and wave sheltered sites; 2) the recruits of the primary layer differed in their impact on the epidermal cell anatomy for most seagrass leaves; 3) the kinds of epiphytes differed

132

------between wave exposed and sheltered sites; and 4) the attachment structures of epiphytes to leaves were heterogeneous in morphology and polysaccharide chemistry. The modular, replicate nature of seagrass leaves and the rapid life histories of the colonists make this system ideal for the study of fundamental ecological processes in communities. While patterns in epiphyte diversity and distribution on seagrass leaves have been considered previously, the role of fine-level interspecies interactions within these near-shore benthic communities have rarely been considered and are unknown for the Hawaiian seagrass and epiphyte systems. Crustose coralline red algae were the most common organisms attached directly to the seagrass leaves at the wave exposed site. The coralline cells were commonly multilayered and underlaid diatoms, filamentous red algae, cyanobacteria and bacteria. The overall community morphology may be considered as a "palisading" of cell layers that has been reported to facilitate solute transport within microbial films (Costerton et ale 1981). The proximity of the cells are suggestive of potential metabolic interactions, although we do not know the physiological nature of interspecies interactions in this system. While the epiphytic organisms nearest the host cells were primarily photoautotrophs at the wave exposed site,

133 the organisms nearest the host cells from the wave sheltered site were primarily bacteria, presumably heterotrophs. The number of cell layers was less at this site and the bacteria that were closely associated with the leaf surfaces were also within the cell wall and adjacent epiphytes. Crustose coralline red algae are cited as the first colonists on seagrasses in other systems (van den Ende and Haage 1963, HUmID 1964, Bramwell and Woelkerling 1984) and develop to reproductive maturity rapidly (Borowitzka and Lethbridge 1989). structural pathways have been reported between a pioneer coralline alga and a seagrass (Ducker and Knox 1978, 1984), but no such intercellular connections were found between Halophila hawaiiana and its algal colonists. This non-invasive mode of epiphytism on H. hawaiiana is termed holoepiphytism following the attachment system based classification of Ducker and Knox (1984). From the current study, it is clear that distinctive modifications of the leaf epidermis occurred at the level of the associated species rather than the level of intrusive connecting structures. The significance of the modification may be in the reported function of the observed cellular structures. The ingrowth region, reported to be important in intensive transport (Barnabas 1988) and osmoregulation in other seagrasses (Jagels 1973, 1983), is elaborately developed adjacent to the algal encrusted areas. The surface area of the wall ingrowths in leaves of H. hawaiiana

134 was greatest in areas under crustose coralline algae. In a study of the seagrass, Phyllospadix, Harlin (1971) noted the possibility that the wall ingrowth region may be involved in the transfer of nutrients and photosynthates to epiphytic algae. No modification of this area was noted however, in epiphytized cells. Other researchers have speculated that the ingrowth structure may be augmented in response to 1) adverse leaf area to volume relationships between source cells and sink cells in putative transport pathway and 2) potential transported solutes that are accompanied by a minimal flow of solvent. Two possible outcomes might be expected from settlement by coralline algae on leaves of seagrasses. First, the encrustation of the leaf with the calcium carbonate cell wall of the crustose coralline algae could hinder metabolic processes of the leaf. At least one study reports a decrease in host wall permeability as well as light penetration following recruitment by crustose coralline algae (Jagels 1983). Alternatively, modifications of the epidermal cells observed in this study indicate that no shade acclimation occurred as algae encrust leaves of H. hawaiiana. The elaborate ingrowths included chloroplasts and mitochondria in approximately similar number and configuration as in unepiphytized leaves. Secretory organelles and their

135 products were more abundant near the regions of elaborate cell wall ingrowths under encrusting epiphytes than elsewhere. This could signal that some functions of the leaf may be specifically modified in response to one type of algae recruiting on the leaf. possible physiological modifications in addition to ingrowth elaboration include assembly, modification and deposition of complex polysaccharides and cell wall glycoproteins for extracellular secretions. Distinctive striations in the leaf cuticle occur between the algal and seagrass cells. These may allow an apoplastic metabolic continuum between the seagrass and algal cells. As discussed in Chapter 2, apoplastic transport between tissues occurs without intercellular connections and is characteristic of many aquatic plants. In addition to a role in host cell modification, encrustation by coralline algae may affect recruitment by other colonists. The primary space held by the corallines diminishes the available space for colonization, and the majority of their cell walls were not colonized. Corallines may be an effective barrier to colonization of the primary leaf surface by other organisms in the 12 day cycle studied here. Diatoms were common colonists on H. hawaiiana from both sites. Diatoms in general are early colonizers of seagrasses and form particularly strong attachments to leaf

136 surfaces (Tanaka 1986). Nutrient dynamics of an attached microbial consortia may be subsidized by release of extracellular material from attached diatom cells that then support associated bacteria (Characklis and Cooksey 1983, Murray and Wetzel 1987). In this study, colonization of leaves from both sites by pennate diatoms did not alter the seagrass anatomy. However, striations in the cuticle may be pronounced under epiphytic diatoms as they are under crustose coralline algae. Only two cyanobacteria types were observed in this study. Cyanobacterial associations with seagrasses have been reported to greatly increase ecosystem production in nutrient-poor environments via nitrogen fixation (Goering and Parker 1972, Capone and Taylor 1980). Although this finding has been disputed (McRoy et ale 1973, Patriquin and Knowles 1972), heterocystic cyanobacteria have been found on leaves as well as in the rhizosphere of seagrasses (Capone and Taylor 1980, Capone 1983). Epiphytic bacteria may be able to fix nitrogen and make it available to the host and other epiphytes without penetration of the host tissues (Kuo 1978). Although heterocysts were not present in the species on H. hawaiiana, the role of cyanobacteria in nitrogen cannot be discounted without a more complete study of different aged leaves and other factors.

~lliile a diverse array of microbes was illustrated, it was beyond the scope of this study to calculate bacterial

137 diversity because of difficulties in distinguishing bacterial species. Environmental influences and developmental stages influence the morphologies of bacterial cells (Moriarty and Hayward 1982, Beveridge 1989). Few studies undertake bacterial taxonomy on anything less than pure cultures growing under defined conditions. The potential for interaction 1) within microbial consortia and 2) between microbes and substrates has been emphasized in the field of biofouling. The many interactions between bacterial cells define the consortia in ways that cannot be understood by the study of solitary cell. The interactions can lead to syntrophic relationships (Cooksey and Cooksey 1986), sYmbiosis with regard to space and substrate and characteristics of molecular diffusion within the microenvironment of the microbes (Wanner and Gujer 1986). Diverse bacteria were clearly present on the seagrass leaves in this study. structural relationships of microbes are evident here, the physiology of the microbes is yet to be determined. Aspects of cells can be interpreted to reveal physiological states, even though these data are structural. Evidence for diverse physiological states of bacteria in the epiphytic consortia includes heterogeneous EPS, granular storage inclusion and "void" areas around the bacterial cells. Those components occurred in epiphytic microbes that were common in this study. Examples of physiological states

138 that may be indicated by those structures include: 1) the bacteria with clear zones and dense polysaccharide may be in states of low metabolic activity characteristic of completed slime production (Jones et ale 1969); 2) capsule elaboration can be induced by external stimuli and sometimes by a particular carbon source (Decho 1990); 3) and storage granules, putative carbon reserves, are regarded as the direct result of the nutritive input from the growth environment (Duguid and Wilkenson 1961). High carbon and low nitrogen, sUlfur, or phosphorus regimes are conducive to their formation (Shively 1974, Preiss 1989). The precise relationship of bacteria to a host differs in the position of the microbial cells to the protoplasm. The relationship of abundant solitary bacteria to H. hawaiiana, and bacteria within the leaf's cell wall were consistently associated with cytological modifications. Modifications included 1) disruption of cell wall fibrils and lysed areas adjacent to invasive cells, 2) diminished wall ingrowths 3) distention into the cytoplasm by secreted material between the cell wall and plasmalemma, 4) membrane bound inclusions, 5) abundant electron-dense spheres, and 6) reduction of mitochondria and chloroplast number. Bacteria attached directly to the leaves induced senescence and cell wall lysis. Colonization of leaves by epiphytes has been considered an important evolutionary pressure, driving growth

139 strategies of seagrasses to increase both the rate of leaf turnover and the rate of withdrawal of nutrients from senescing leaves (Josselyn et ale 1986, Hemminga et al. 1991). Periodic or continual shedding of leaves in some seagrass species (including Halophila) may minimize the shading effects on seagrass photosynthesis (Sand-Jensen 1977, Borum and Wium-Anderson 1980, Bulthuis and Woelkerling 1983, Borowitzka and Lethbridge 1989). The results presented here indicate that not all epiphytes alter the leaf epidermis. Cytological modifications appear to be specific and consistent for each epiphytic taxon studied. In contrast to the effects of solitary bacteria on the leaves, bacteria in microcolonies do not appear to be either directly attached to the host or to cause cytological modifications. The invasive properties of bacteria appear to differ between solitary bacteria and bacteria in microcolonies. The composition of bacterial EPS are heterogeneous among the class of solitary bacteria, and may differ from that of bacteria in microcolonies. The microbial EPS are key to the ecology of the microbial cell as well as attachment to a substratum. The EPS create a micro-environment around the cell that important in efficient metabolism and reproduction and bUffering the cells from rapid changes in ion, pH, salinity, desiccation or nutrient regimes (Boyle and Reade 1983). A

140

- . ------bacterial cell can also sequester and concentrate nutrients in the EPS (Costerton et ale 1981). The external cell coat provides a relatively constant molecular environment close to the cell wall and the coating layer holds or retains degradative enzymes needed to hydrolyze polymeric substrates. The bacterial glycocalyces observed in this study appear to be diverse in terms of carbohydrate chemistry with respect to attachment. Neutral polysaccharides, acidic polysaccharides and protein associated polysaccharides were each found in attachment regions of bacteria. Even though the specific chemistry of attachment remains undetermined, general classes of polysaccharides were localized by specific probes used in this stUdy. Dense staining of glycocalyx components (without tannic acid and/or Ruthenium red) indicated presence of a protein within the glycocalyx as well as carbohydrate; glycoproteins may be components of specific cell surface mucilages. The dense staining of bacterial glycocalyces by tannic acid identified extracellular mucilage to be rich in complex carbohydrates inclUding neutral polysaccharides (Sannes et ale 1978) and may localize glycosaminoglycans (Singley and Solursh 1980). Ruthenium red localized acidic polysaccharides were common components of the bacterial glycocalyces inclUding the common matrix of bacteria in microcolonies. Diverse

141 attachments were evident between cells including microcolonies and attachments of epiphytes to hosts. The environmental conditions present across the wave force gradient examined in this study seems likely to have influenced attachments of epiphytes to hosts at several levels. The forces of attachment may vary according to the hydrodynamic conditions. Particles including propagules or microorganisms are transported by fluid dynamic forces to the point where van der Waals (electrodynamic) forces act upon cells near a surface: the interplay of physical forces have been demonstrated in some plant/bacterial associations (van Loosdrecht et ale 1989). Hydrodynamic shear forces are particularly important in initial, reversible adherence of a cell to a substratum (Marshall et ale 1971a, van Loosdrecht et ale 1989) as well as the production of microbial EPS. Water movement determines dispersal and affects rates of recruitment in seagrass meadows (Eckman 1983). All colonizers are delivered to substrates via current direction and velocity (Korte and Blinn 1983, Roughgarden et ale 1986). This is reflected as variations in species diversity, micro-patterns in biofilm topography, and rates of cellular accretion. Distribution of algal epiphytes on seagrass leaves (Harlin 1971), as well as the cell layers and density have been found to vary as a function of water velocity (Korte and Blinn 1983). The distinctive patterns in site specific recruitment in this study occur in both the

142 morphologies of the epiphytes and the species composition. Leaves reduce current speeds and propagules are retained by the seagrass populations. This baffle effect enhances recruitment rates by increasing the "window of opportunity" for settlement (Hootsmans and Vermaat 1983, Wilson 1990). The sediment composition is also affected by this process (Eckman 1983). The differences in epiphytes on H. hawaiiana leaves from the wave exposed and sheltered sites may be attributable, in part, to the different leaf densities per site. Not only are leaves sparse in wave exposed areas, but patches are ephemeral with borders that shifted throughout the year; leaves are even covered by sand occasionally. The sediment underlying the seagrass was similar to adjacent areas without vegetation. Removal of propagules and sediment from the water column may be minimal at this wave exposed site yet the greater water motion appears to deliver adequate propagules for epiphytic colonization. The leaf abundance from the wave sheltered site was dense. The patches appeared constant throughout the year; suspended particles in the water column were commonly observed yet the leaves were not covered over by sediment as occurred in the wave exposed site. The sediment underlying the seagrass differed from adjacent areas without vegetation; a deep anoxic layer immediately underlaid leaves. Removal of propagules and sediment from the water

143 column potentially would be great here because of the high leaf density. The minimal water flow could minimize nutrient delivery to the epiphytes but a higher ambient nutrient regime at this site may mitigate this factor. The distinctive differences in recruitment on H. hawaiiana leaves in these wave exposed and sheltered sites indicate that the pool of propagules may have differed as well as the microenvironment for settlement. Additionally, propagule differences such as the rapidity of attachment and strength of attachment may influence the ultimate composition of an epiphyte flora across a wave force gradient as in this study. Other features, such as spore diameter or weight, may facilitate their retention in the boundary layer of seagrass leaves (Boney 1975). Reproductive processes that generate algal propagules are poorly understood as is the ecological importance of positive interspecific and intraspecific interaction in settlement and recruitment. The attachment abilities of benthic algae change with time and species; dispersal as the scattering of propagules in all directions from the parental thalli is an individual rather than a collective process which depends on the characteristics of the plants as well as on the environment. These features are the result of several variables including the type and dispersability of the unit and concentration at the source (Santelices 1990). We do not know the dispersal characteristics of the parental

144 propagule stock at Maleakahana and Moku'auia or the impact

of site specific environmental parameters on the dispersal

process, only the resultant recruitment patterns illustrated

in this study.

conclusions The general value of seagrasses in near shore

ecosystems has been referred to as a potentially tremendous

source of primary productivity a substrate for epiphytes

that augment local primary production an'd add diversity to

those ecosystems. The results of this study indicate that

distinct biotic and abiotic factors determine the nature of

the seagrass and epiphyte community. The general null hypothesis that wave exposure does not influence

colonization of seagrass leaves by epiphytes is disproven within the constraints of this study. While similar taxa of

epiphytes occurred within the sites, the primary colonists,

or organisms adnate to the host, differed. The different

taxonomic categories of epiphytes that dominated at the wave

exposed or sheltered sites were associated with a range of

anatomical modifications of the host. Diatoms,

cyanobacteria and bacteria in microcolonies or microbes

associated but not attached to the leaves did not alter the

host anatomy. However, the epidermal areas adjacent to

solitary bacteria (particularly those within the seagrass

cell wall) and crustose coralline red algae were

145 consistently modified in dissimilar ways. Those epiphytes were attached directly to the leaves by cell to cell adhesion that did not involve intrusive structures; interspecific cell to cell interactions were carried out at the cell surfaces. The dominant epiphyte that adhered directly to the host appeared to influence subsequent colonization of the leaves. The host and epiphyte interface contained a number of diverse extracellular polysaccharides from microbes. Cells attached to the leaves by both neutral and acidic extracellular polymers. This diminutive benthic community was composed of species in a range of complex and specific interactions that, in some cases, appear to determine the structure of the leaf and epiphyte species.

146 Table 3.1. Preservation of EUkaryotic and Prokaryotic Cells and Cell Products

COMPONENT FIXATION A FIXATION ~ FIXATION £ FIXATION ~ FIXATION E. ..!.BBl (Lysine) ~ (TA+RR) i. hawaiiana protoplast +++ ++ ++ ++ ++ cell wall ++ ++ ++ ++ crustose cora.lline ++ ++ ++ ++ ++ algae filamentous red algae ++ ++ NE NE NE

...... diatom ~ -..J protoplast ++ ++ ++ ++ ++ organic layer ++ ++ ++ ++ ++ blue-green lbac:1;eria protoplast ++ ++ ++ ++ ++ glycocalyx ++ ++ ++ ++ ++ bacteria protoplast ++ ++ ++ ++ ++ glycocalyx ++ ++ ++ ++

Membrane clearness and overall appearance was assessed on a scale of - to +++i - = disruption of cell components, + = information attainable but resolution poor or inconsistent throughout tissue, ++ = consistently good cellular details, +++ consistant, sharp resolution, preferred fixation. Table 3.2. A profile of Halophila hawaiiana and its epiphytes; number of epiphyte cells/host cell. N = 100 host cells site ALGAE BACTERIA cc fr dt as sol at sol 22 mic at mic sa invasive Ma1eakahana, wave exposed site-- unepiphytized leaves (less than 10% cover) % present 3 12 15 27 12 6 5 0 mean 2.7 2.3 1.6 3.1 2.5 4.2 4.2 5.2 S.D. 0.6 0.6 0.7 2.4 1.0 1.8 1.0 1.9 epiphytized leaves (ca. 25% cover) % present 72 9 19 11 89 1 27 0 mean 5.3 0.1 2.2 3.0 1.9 .0 1.0 9.7 S.D. 2.7 2.0 0.9 1.4 1.1 7.0 .0 4.9

Hoku'auia, wave sheltered site .... .&:>0 ce unepiphytized leaves (less than 10% cover) % present 2 5 9 10 19 83 6 12 6 mean 3.5 2.4 1.7 2.2 1.7 2.6 2.8 3.4 1.5 S.D 0.7 0.5 0.7 0.8 0.7 1.2 1.2 0.9 0.5 epiphytized leaves (ca. 25% cover) % present 8 5 9 23 27 100 8 35 54 mean 2.2 0.2 1.8 2.1 2.0 8.6 3.2 5.6 5.6 S.D. 0.7 0.4 0.7 0.9 1.1 6.4 1.8 1.7 5.4

ALGAE BACTERIA cc = crustose coralline algae, rhodophyta bg = blue-green bacteria, cyanobacteria fr = filamentous red algae, rhodophyta sol at = attached solitary cells dt = diatoms, chrysophyta sol as = associated solitary cells mic at = attached cells in microcolony mic as = associated cells in microcolony invasive = within cell wall or cytoplasm Table 3.3. A profile of Halophila hawaiiana leaf cell components; number of chloroplasts and layers of cell wall and plasmalemma elaborations N = 100 host cells site CHLOROPLASTS WALL ELABORATION LAYERS Malaekahana unepiphytized leaves (less than 5% cover) % present 100 mean 9.4 1.0 S.D. 1.9 0.0 epiphytized leaves (ca. 25%) % present 100 mean 9.2 1.7 S.D. 1.3 0.7

Moku'auia unepiphytized leaves (less than 5% cover) % present 100 mean 11.1 0.0 S.D. 1.9 0.0 epiphytized leaves (ca. 25%) % present 100 mean 8.9 0.9 S.D. 1.3 0.3

149 Table 3.4. Assessment of significant differences within categories (99% confidence interval). ~ =highest mean, ~ =lowest mean. CATEGORY GROUP VARIABLE crustose coralline algae B unepiphytized, exposed A epiphytized, exposed B unepiphytized, sheltered B epiphytized, sheltered solitary bacteria -attached to seagrass B unepiphytized, exposed B epiphytized, exposed B unepiphytized, sheltered A epiphytized, sheltered solitary bacteria -associated with seagrass C unepiphytized, exposed A epiphytized, exposed B unepiphytized, sheltered A epiphytized, sheltered bacteria in microcolonies -associated with seagrass B unepiphytized, exposed A epiphytized, exposed B unepiphytized, sheltered A epiphytized, sheltered bacteria in cell wall or cytoplasm (invasive) B unepiphytized, exposed B epiphytized, exposed B unepiphytized, sheltered A epiphytized, sheltered chloroplasts B unepiphytized, exposed C epiphytized, exposed A unepiphytized, sheltered D epiphytized, sheltered wall ingrowth layers B unepiphytized, exposed A epiphytized, exposed B unepiphytized, sheltered C epiphytized, sheltered exposed = wave exposed, Malaekahana sheltered = wave sheltered, Moku'auia

150 Figure 3.1. Multilayered epiphyte community on adaxial seagrass leaf, wave exposed site. I. A bacterial microcolony (me) and glycogen-like inclusions (arrow), cyanobacterial cells (cb) and crustose coralline red algal cells (cc). TEM. Fixation A. Scale = 0.8 ~m. Figure 3.2. MUltilayered epiphyte community on adaxial seagrass leaf, wave exposed site. II. Diverse bacteria (b) and glycocalyx fringed cell ghosts (g) associated with filamentous red algal cells (ra) that overlay crustose coralline red algal cells (cc) adnate to seagrass leaf. Cytoplasmic inclusion, poly-~-hydroxybutyrate-likegranule in epiphytic bacterium (arrow). TEM. Fixation A. Scale = 4.0 ~m.

Figure 3.3. Multilayered epiphyte community on adaxial seagrass leaf, wave exposed site III. Multiple layers of crustose coralline red algal cells (cc) encrust seagrass leaf, epidermal cell in oblique cross section. TEM. Fixation A. Scale = 4.0 ~m. Figure 3.4. MUltilayered epiphyte community on adaxial seagrass leaf, wave exposed site IV. Multiple layers of crustose coralline red algal cells (cc) encrust seagrass leaf, elaborate ingrowth region (I) of epidermal cell wall/cytoplasm. TEM. Fixation A. Scale = 2.0 ~m.

151 152

----- Figure 3.5. Multilayered epiphyte community on adaxial seagrass leaf, wave exposed site V. An organic coating of a pennate diatom adheres frustule to crustose coralline red algal cells (cc) adnate to seagrass. Cells of a cyanobacterium (cb) and bacteria with poly-~ hydroxybutyrate-like inclusions (arrow) are embedded within the matrix. TEM. Fixation A. Scale = 0.8 ~m. Figure 3.6. Pennate diatoms adherent to seagrass leaf, wave exposed site. TEN. Fixation B. Scale = 1.6 ~m. Figure 3.7. Inset: outer cytoplasm of unepiphytized epidermal cell, wave exposed site. Cell wall ingrowth region similar in epiphytized leaf and diatom colonized leaf. TEM. Fixation A. Scale = 2.0 ~m. Figure 3.8. A diatom frustule remains attached to the seagrass leaf by an apparent attachment plug (pl) following death of the cell. Fixation B. Scale = 0.13 ~m.

153 ~./

.~~.

154 Figure 3.9. Unepiphytized adaxial seagrass leaf, a control for assessing cellular modification associated with epiphytes. Fibrous inner wall (iw), electron dense ingrowths (I), invaginated plasmalemma (arrow) with mitochondria (m) and ribosomes (r) associated. TEM. Fixation A. Scale = 0.25 ~m. Figure 3.10. Epidermal cell modifications associated with coralline red algal encrustation (cc), wave exposed site. I. Ingrowth region becomes dense with secretory vesicles as well as chloroplasts (cl) and mitochondria (m). Striations in the seagrass cuticle evident (arrow). TEM. Fixation A. Scale = 0.5 ~m. Figure 3.11. Epidermal cell modifications associated with coralline red algal encrustation, wave exposed site. II. Dictyosome (arrow) with apparent forming face to cell wall ingrowth (I) and dense material in apoplastic space between ingrowth and inner cell wall (iw). TEM. Fixation A. Scale = 0.19 ~m. Figure 3.12. Epidermal cell modifications associated with coralline red algal encrustation, wave exposed site. III. Elaborate extensions of ingrowths (I) into cytoplasm near mitochondria (m). striations in cuticle evident (arrow). TEM. Fixation B. Scale = 0.3 ~m.

155 156 Figure 3.13. Epidermal cell modifications associated with bacteria within cell wall, wave exposed site. Diatoms (d) not associated with modifications of the protoplasts. Invasive bacteria (arrow) associated with distended ingrowth (arrowhead). TEM. Fixation B. Scale = 5.0 ~m. Figure 3.14. Higher magnification of bacterium within seagrass cell wall, wave exposed site I. Ingrowths (I) appear coalescent with secretory material distended into cytoplasm (double arrows). A halo of altered cell wall surrounds bacterial cell (arrow). TEM. Fixation B. Scale = 0.6 ~m.

157

....__ . __._------.~ 3.13

158 Figure 3.15. Overview of bacteria on leaf, wave exposed site. Upright (u) and prostrate (p) rod shaped bacteria and a Ilhemisphericalll bacterium with apparent attachment fringe (arrow) and scattered organic matter (om). SEM. Fixation A. Scale = 2.5~m. Figure 3.16. variation in glycocalyces of bacteria attached to seagrass leaves at wave exposed site. I. Bacteria adhere by contrasting polysaccharides. Homogeneous fibrillar polysaccharide (arrow) contrasts with electron dense adhesive fringe (f). Poly-p-hydroxybutyrate-like granules (h) may occlude nucleoplasm, wave exposed site. TEM. Fixation C. Scale = 0.4 ~m. Figure 3.17. variation in glycocalyces of bacteria attached to seagrass leaves at wave exposed site. II. A non-staining or void area (vo) around the cell is adjacent to a capsule with lighter stain response than the surrounding mucilage. Internal membranes are evident (arrow). TEM. Fixation E. Scale = 0.3 ~m. Figure 3.18. variation in glycocalyces of bacteria attached to seagrass leaves at wave exposed site. III. Mucilage attaches the basal cell of a cyanobacterium to the seagrass leaf (arrow). Polysaccharide capsule and void (vo) area surrounding cells are evident. A capsule is tightly adherent to the cell wall. TEM. Fixation E. Scale = 0.5 ~m

159 160 Figure 3.19. variation in glycocalyces of bacteria associated with seagrass leaves at wave exposed site. I. Coccoid bacteria in leaf "slime layer". Fibrillar nature of the polysaccharide (arrow). Bacterium (upper left) within extended capsule (ca) with nucleoplasm (n). TEM. Fixation A. Scale = 0.25 ~m. Figure 3.20. variation in glycocalyces of bacteria associated with seagrass leaves at wave exposed site. II. Coccoid bacteria in leaf slime layer with nucleoplasm (n), bacterial wall (w) and glycogen-like granule (arrow). Clay/silt particles within matrix (cp). TEM. Fixation E. Scale = 5.0 ~m. Figure 3.21. variation in glycocalyces of bacteria associated with seagrass leaves at wave exposed site. III. Coccoid bacterium with polar distribution of tannic acid positive mucopolysaccharide (arrow). Membrane doublets concentric in cell (double arrows). clay/silt particles within matrix (cp). TEM. Fixation E. Scale = 5.0 ~m.

Figure 3.22. Variation in glycocalyces of bacteria associated with seagrass leaves at wave exposed site. IV. Fimbria-like extensions from coccoid bacterial cell wall (arrow) associated with tannic acid positive mucilage (mu). TEM. Fixation E. Scale = 4.0 ~m. Figure 3.23. Variation in glycocalyces of bacteria associated with seagrass leaves at wave exposed site. V. Microcolony of rod-shaped bacteria. Tangential section through microcolony of rod-shaped bacteria above cyanobacterium (arrow), part of multilayered community in Fig. 3.1. Glycogen-like granules (arrowhead) within central nucleoplasm. Clay/silt within the microcolony matrix (me). TEM. Fixation A. Scale = 1.0 ~m. Figure 3.24. Bacteria associated with the seagrass leaves, variation in glycocalyces, wave exposed site VI. Irregularly shaped bacteria in microcolony fringed by clay silt-debris adherent to glycocalyx. Glycogen-like granules (arrow) and poly-~-hydroxybutyrate-likeinclusions (h) in the bacteria. TEM. Fixation A. Scale = 0.6 ~m.

161 f/f:"'C ,'-'.' .' :.~' ••.' If • ••••C •• .

162 Figure 3.25. Epiphyte community on adaxial seagrass leaf, wave sheltered site. I. Overview of epidermal cells heavily epiphytized with bacteria. Vesicles evident between the invaginated and inner cell wall (arrow), no mitochondria, condensed chloroplasts (cl), electron dense spheres (doUble arrow) and unidentified membrane-bound inclusions (0). Bacteria (b) and cyanobacteria (cb) adherent to and associated with the seagrass leaf. TEM. Fixation D. Scale = 0.8 J.Lm. Figure 3.26. Epiphyte community on adaxial seagrass leaf, wave sheltered site. II. Multilayered community includes tunneling bacteria (t) and diatom frustule with bacteria (arrow) under, crustose coralline red algae (cc) and surface bacteria. Electron dense material adjoins electron dense spheres, probably lipid (arrowhead). TEM. Fixation C. Scale = 2.0 J.Lm. Figure 3.27. Epiphyte community on adaxial seagrass leaf, wave sheltered site III. Crustose coralline red algae (cc) overlay bacteria infected diatom (d), adnate to separating seagrass cuticle (arrow). Cyanobacterium in matrix of alga (cb). Fixation C. TEM. Scale = 2.0 J.Lm. Figure 3.28. Epiphyte community on adaxial seagrass leaf, wave sheltered site IV. Bacteria (b) within the epiphytic diatoms frustule (d), on the seagrass leaf and within the cell wall. Bacterial cell ghosts (g) with intact glycocalyces overlay the primary leaf colonists. TEM. Fixation C. Scale = 1.0 J.Lm. Figure 3.29. Epiphyte community on adaxial seagrass leaf, wave sheltered site V. Crustose coralline algal cells (cc) overlaid diatoms (d) and bacteria (b) that were adherent to the seagrass cuticle. Bacterial with FHB-like inclusions (arrow) overlay apparent secreted products from an adjacent diatom (not pictured). Ensheathed cyanobacterial cells (cb), bacteria and clay (cp) were adjacent to the coralline algal cell wall. Bacteria (t) were in cell wall. TEM. Fixation C. Scale = 2.0 J.Lm

163

-----._------.t'boe ".C' .... ~ .G

.~ .

164 Figure 3.30. Diatom adherent to the seagrass, carbohydrate containing material adjoins diatom to leaf. striations in cuticle were evident near lysed area of the cell wall (arrow). Bacteria occurred at lysed leaf (b), wave sheltered site. TEM. Fixation B. Scale = 0.8 ~m. Figure 3.31. Profile of epiphyte and seagrass interface. Bacteria colonize surface diatom and seagrass, wave sheltered site. I. Striations in cuticle underlay diatom frustule (arrow). Bacteria in diatom frustule and outer cell wall of seagrass. Electron dense spheres (sp) are adjacent to cell wall ingrowths (I). TEM. Fixation c. Scale = 0.5 ~m. Figure 3.32. Profile of epiphyte and seagrass interface. Bacteria colonize surface diatom and seagrass, wave sheltered site. II. Extracellular material (mu) and a bacterium (arrow) between diatom and seagrass leaf. TEM. Fixation C. Scale = 0.4 ~m. Figure 3.33. Profile of epiphyte and seagrass interface. Bacteria colonize surface diatom and seagrass, wave sheltered site. III. Extracellular material between a diatom and bacterium, and the seagrass leaf (mu) and metabolic products of tunneling bacteria within the seagrass cell wall (arrow). TEM. Fixation C. Scale = 0.8 ~m.

165 166 Figure 3.34. Overview of bacteria on seagrass leaf, wave sheltered site. Coccoid bacteria with pili (arrow), rod shaped bacteria (b) and "hemispherical bacteria" (arrowhead) with attachment fringe are adnate to the seagrass leaf, organic matter present (om). SEM. Fixation A. Scale = 2.5 p.m.

Figure 3.35. Hemispherical bacterium adnate to the seagrass leaf. Poly-p-hydroxybutyrate-like granules (h) within nucleoplasm area, wave sheltered site. TEM. Fixation C. Scale = 0.25 J.Lm. Figure 3.36. Coccoid bacterium adherent to the seagrass by fibrous glycocalyx (ca). Doublets of internal membranes (arrow) in electron dense cytoplasm. Tannic acid positive mucilage at glycocalyx fringe (mu), wave sheltered site. TEM. Fixation E. Scale = 0.19 p.m. Figure 3.37. Bacteria with heterogeneous glycocalyces (arrows) attach to or associated with degrading seagrass wall, wave sheltered site. TEM. Fixation D. Scale = 0.25 J.Lm. Figure 3.38. Bacterial infection at lysed seagrass leaf, wave sheltered site. Cuticle fragment remained (arrow) and abundant bacteria with poly-p-hydroxybutyrate granule-like inclusions (arrowhead). TEM. Fixation C. Scale = 2.0 J.Lm.

167 168 Figure 3.39. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. I. Solitary bacterium within void area (arrow) surrounded by amorphous mucilage that is positive to tannic acid (mu). Cell wall matrix (ow) tannic acid positive and cuticle lacking, wave sheltered site. TEM. Fixation D. Scale = 0.19 ~m. Figure 3.40. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. II. Compact glycocalyces (arrow) adnate to bacterial cell wall anchor cells to seagrass leaves. TEM. Fixation D. Scale = 0.25 ~m. Figure 3.41. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. III. Dividing bacterium bounded by dense mucilage (arrow). Poly-~-hydroxybutyrate-likegranules (h) adjacent to electron dense invaginations (I) of the bacterial wall. cytokinesis stage of cell division. TEM. Fixation D. Scale = 0.19 ~m. Figure 3.42. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. IV. Dividing bacterium with expanded nucleoplasm (n) adjacent to bacterial wall invaginations. Cell bounded by fibrillar polysaccharide (ca) and dense mucilage. TEM. Fixation D. Scale = 0.19 ~m. Figure 3.43. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. V. Dividing bacterium, early stage of cytokinesis with dense mucilage (arrow) adnate to bacterial cell wall. TEM. Fixation D. Scale = 0.25 ~m. Figure 3.44. Diverse bacteria attached to seagrass leaf with distinctive glycocalyx components, wave sheltered site. VI. Dense tannic acid mucilage surrounds void area (arrow) around bacterium and contrasts with fibrous polysaccharide that attaches the cell to the seagrass. TEM. Fixation D. Scale = 0.19 ~m.

169 , =- ~ -.~.- -~ -- --~ -~~ ,or' ----:--- ~~~+--~- --~ -- ~-=----~------~-

170 Figure 3.45. Diverse bacteria associated with seagrass leaves, wave sheltered site. I. cyanobacterial trichome (cb) near abaxial leaf surface (arrow) (only outer cytoplasm presented due to poor infiltration). No direct cellular attachment observed. TEM. Fixation B. Scale = 4.0 ~m. Figure 3.46. Diverse bacteria associated with seagrass leaves, wave sheltered site. II. Filamentous bacterium with heterogeneous cell wall that shows areas positive to tannic acid (arrow). TEM. Fixation D. Scale = 0.25 ~m. Figure 3.47. Diverse bacteria associated with seagrass leaves, wave sheltered site. III. Bacteria cells aggregate to mucilage that is positive to tannic acid (arrow). TEM. Fixation D. Scale = 0.25 ~m. Figure 3.48. Diverse bacteria associated with seagrass leaves, wave sheltered site. IV. Coccoid bacterium with heterogeneous glycocalyx (arrow). TEM. Fixation E. Scale = 0.25 ~m. Figure 3.49. Diverse bacteria associated with seagrass leaves, wave sheltered site. V. Coccoid bacteria with heterogeneous glycocalyces, areas positive to tannic acid (arrow). TEM. Fixation D. Scale = 0.25 ~m.

Figure 3.50. Diverse bacteria associated with seagrass leaves, wave sheltered site. VI. Three adjacent microcolonies (m1, m2, m3) within polysaccharide matrices. Microcolonies 1 and 2 have similar cells in a fibrillar matrix. Void (arrow) around cells. Nucleoplasm, and cytoplasmic inclusions include glycogen-like granules (arrowhead), poly-~-hydroxybutyrate-likegranules (h). The third microcolony contained smaller more irregularly shaped cells with a less distinct matrix. clay/silt accumulations occurred at the microcolony peripheries and around solitary bacteria adjacent to the seagrass leaf (lower right). TEM. Fixation B. Scale = 0.8 ~m.

171 •

.';":.' 3.48

, •pt' ._.. ,. .~-

172 Literature Cited

Akin, D. E., Rigsby, L. L. (1990). Preservation of ruminal bacterium capsules by using lysine in the electron microscopy fixation. Appl. Environ. Microbiol. 56(9):2933-2935. Ballantine, D. L. (1979). The distribution of algal epiphytes on macrophyte hosts offshore from La Parguera, Puerto Rico. Bot. Mar. 22:107-111. Barnabas, A. D. (1988). Apoplastic tracers studied in the leaves of a seagrass. I. Pathway through epidermal and mesophyll tissues. Aquat. Bot. 32:63-77. Beveridge, T. J. (1989). The structure of bacteria. In: Poindexter, J. S., Leadbetter, E. R. (eds.) Bacteria in nature vol. 3; structure, physiology, and genetic adaptability. Plenum Press, New York. Blanquet, P. R. (1976). Ultrahistochemical study on the ruthenium red surface staining. II Nature and affinity of the electron dense marker. Histochemistry 47:175. Boney, A. D. (1975). Mucilage sheaths of spores or red algae. J. Mar. BioI. Ass. U.K. 55:511-518. Borowitzka, M. A., Lethbridge, R. C. (1989). Seagrass epiphytes. In: Larkum, A. W. D., McComb, A. J., Shepherd, S. A. (eds.) Biology of seagrasses. Elsevier Sci. Pub. Co., New York.

Borum, J., Wium-Andersen, S. (1980). Biomass and production of epiphytes on eelgrass (Zostera marina L.) in the Oresund, Denmark. Ophelia, Supple 1:57-64. Boyle, C. D., Reade, A. E. (1983). Characterization of two extracellular polysaccharides from marine bacteria. Appl. Environ. Microbiol. 46:392-399. Bramwell, M. D., Woelkerling, W. J. (1984). Studies on the distribution of Pneophyllum-Fosliella plants (Corallinaceae, Rhodophyta) on leaves of the seagrass Amphibolis antarctica (Cymodociaceae). Aust. J. Bot. 32:131-137. Bulthuis, D. A., Woelkerling, W. J. (1983). Biomass accumulation and shading effects of epiphytes on leaves of the seagrass Heterozostera tasmanica in Victoria, Australia. Aquat. Bot. 16:137-148.

173 Burkholder, P. R. (1973). The ecology of marine antibiotics and coral reefs. In: Jones, O. A., Endean, R. (eds.) Biology and geology of coral reefs 2, biology 1. Academic Press, New York. Brylinsky, M. (1977). Release of dissolved organic matter by some marine macrophytes. Mar. BioI. 39:213-220.

Capone, D. G. (1983). N2 fixation in seagrass communities. Mar. Tech. Soc. J. 17:32-37. Capone, D.G., Taylor, B.F. (1980). Microbial nitrogen cycling in a seagrass community. In: Kennedy, V. S. (ed.) Estuarine perspectives. Academic Press, New York. Characklis, w. G., Cooksey, K. E. (1983). Biofilms and microbial biofouling. Adv. Appl. Microbiol. 29:93-138. Connell, J. H. (1985). The consequences of variation in initial settlement vs. post-settlement mortality in rocky intertidal communities. J. Exp. Mar. BioI. Ecol. 93:11-46. ] Cooksey, K. E., Cooksey, B. (1986). Adhesion of fouling diatoms to surfaces: some biochemistry. In: Evans, L. V., Hoagland, K. D. (eds.) Algal biofouling. Elsevier Science Publishers B. V. Amsterdam. Corpe, w. A. (1980). Microbial surface components involved in adsorption of microorganisms onto surfaces. In: Bitton, G., Marshall, K. C. (eds.) Adsorption of microorganisms to surfaces. John Wiley, New York. Costerton, J. W., Irvin, R. T. (1981). The bacterial glycocalyx in nature and disease. Ann. Rev. Microbiol. 35:299-324. Costerton, J. W., Irvin, R. T., Cheng, K. J. (1981). The bacterial glycocalyx in nature and disease. Annu. Rev Microbiol. 35:299-234. Costerton, J. W., Marrie, T. J., Cheng, K. J. (1985). Phenomenon of bacterial adhesion. In: Savage, D. C., Fletcher, M. (eds.) Bacterial adhesion. Plenum Press, New York. D'Antonio, C. (1985). Epiphytes on the rocky intertidal red alga Rhodomela larix (Turner) C. Agardh: Negative effects on the host and food for herbivores? J. Exp. Mar. BioI. Ecol. 86:197-218.

174 Decho, A. W. (1990). Microbial exopolymer secretions in ocean environments: their role(s) in food webs and marine processes. Oceanogr. Mar. BioI. Annu. Rev. 28:73-153. Dowdy, S., Wearden, S. (1991). statistics for research. John Wiley, New York.

Duarte, C. (1989). Temporal biomass variability and production/biomass relationships of seagrass communities. Mar. Ecol. Prog. Sere 51:269-276.

Ducker, S. C., Knox, R. B. (1978). Alleloparasitism between a seagrass and algae. Naturewissenshaften 65:391-392.

Duguid, J. P., Wilkenson, J. F. (1961). Environmentally induced changes in bacterial morphology. Symp. Soc. Gen. Microbiol. 69-99

Eckman, J. E. (1983). Hydrodynamic processes affecting benthic recruitment. Limnol. Oceanogr. 28:241-257.

Feldman, J. (1937). Recherches sur la vegetation marine de la mediteranee. La Cote des Alberes. Chap. 2. Le substratum. Revue Algol. 10:17-29. Fletcher, M., Floodgate, G. D. (1973). An electron microscopic demonstration of an acidic polysaccharide involved in the adhesion of a marine bacterium to solid surfaces. J. Gen. Microbiol. 74:325-334.

Fletcher, M., Marshall, K. C. (1982). Are solid surfaces of ecological significance to aquatic bacteria? In: Marshall, K. C. (ed.) Advances in microbial ecology. Plenum Press, New York.

Fountaine, T. D., Nigh, D.G. (1983). Characteristics of epiphytic communities on natural and artificial submersed lotic plants. Archiv fur Hydrobiol. 96:293-301.

Goering, J. J., Parker, P. L. (1972). Nitrogen fixation by epiphytes on sea-grasses. Limnol. Oceanogr. 17(2):320 323.

Gunning, B. E. S. (1977). Transfer cells and their roles in transport of solutes in plants. Sci. Prog. 64:539-568.

175 Gunning, B. E. S., Pate, J. S. (1969). "Transfer cells"- the plant cells with wall ingrowths, specialized in relation to short transport of solutes--their occurrence, structure and development. Protoplasma 68:107-133. Harlin, M. M. (1971). Epiphytic marine algae: interactions with their hosts. Ph.D. dissertation, University of Washington. Seattle, Wash. Harlin, M. M. (1973). Transfer of products between epiphytic marine algae and host plants. J. Phycol. 9:243-248. Harlin, M. M. (1975). Epiphyte-host relations in seagrass communities. Aquat. Bot. 1:125-131. Harrison, P. G. (1982). Control of microbial growth and of amphipod grazing by water-soluble compounds from leaves of Zostera marina. Mar. BioI. 67:225-230. Harrison, P. G., Chan, A. T. (1980). Inhibition of the growth of micro-algae and bacteria by extracts of eelgrass (Zostera marina) leaves. Mar. BioI. 61:21-26. Heijs, F.M.L. (1985a). The seasonal distribution and community structure of the epiphytic algae on Thalassia hemprichii (Ehrenb.) Aschers. from Papua New Guinea. Aquat. Bot., 21:295-324. Heijs, F.M.L. (1985b). The macroalgal component in monospecific seagrass beds from Paupua New Guinea. Aquat. Bot. 22:291-324. Hemminga, M. A., Harrison, P. G., Lent, F. v. (1991). The balance of nutrient losses and gains in seagrass meadows. Mar. Ecol. Prog. Sere 71:85-96. Hootsmans, M. J. M., Vermaat, J. E. (1985). The effect of periphyton-grazing by three epifaunal species on the growth of Zostera marina L. under experimental conditions. Aquat. Bot. 22:83-88. Hough, R. A., Wetzel, R. G. (1975). The release of dissolved organic carbon derived from aquatic macrophytes: diel, seasonal, and community relationships. Bern. Int. Ver. Theor. Angew. Limnol. 19:939-948. Hudon, C., Bourget, E. (1981). Initial colonization of artificial substrate: community development and structure studied by scanning electron microscopy. Can. J. Fish. Aquat. Sci. 38:1371-1384.

176 HUmID, H. A. (1964). Epiphytes of the seagrass, Thalassia hemprichii, in Florida. Bull. Mar. Sci. Gulf caribb. 14: 306-341. Jacques, M., Graham, L. (1989). Improved preservation of bacterial capsule for electron microscopy. J. Electron Microsc. Techn. 11:167-169. Jagels, R. (1973). Studies of a marine grass, Thalassia testudinum I. Ultrastructure of the osmoregulatory leaf cells. Am. J. Bot. 60: 1003-1009. Jagels, R. (1983). Further evidence for osmoregulation in epidermal leaf cells of seagrasses. Am. J. Bot. 70:327 333. Jones, H. C., Roth, I. L., Sanders, W. M. (1969). Electron microscopic study of a slime. J. Bacteriol. 99:316-325. Josselyn, M., Fonseca, M., Niesen, T., Larson, R. (1986). Biomass, production and decomposition of a deep water seagrass, Halophila decipiens Ostenf. Aquat. Bot. 25:47 61. Kenworthy, W. J., Currin, C. A., Fonseca, M. S., smith, G. (1989). Production, decomposition, and heterotrophic utilization of the seagrass Halophila decipiens in a submarine canyon. Mar. Ecol. Prog. Sere 51:277-290. Keough, M. J., Downes, B. J. (1982). Recruitment of marine invertebrates: the role of active larval choices and early mortality. oecology 54:348-352. Khan, K.M., Hatfield, J.S. and D.G. Drescher. (1990). The cell coat of the sensory and supporting cells of the rainbow trout saccular macula as demonstrated by reaction with ruthenium red and tannic acid. J. Histochem. Cytochem. 38(11):1615-1623. Kirchman, D. L., Mazzella, L., Alberte, R. S., Mitchell, R. (1984). Bacterial epiphytes on Zostera marina. Mar. Ecol. Prog. Sere 15:207-211. Korte, V. L., Blinn, D. W. (1983). Diatom colonization on artificial substrata in pool and riffle zones studied by light and scanning electron microscopy. J. Phycol. 19:332-342. Kuo, J. (1978). Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae). I. Leaf blade and leaf sheath of Posidonia australis Hook F. Aquat. Bot. 5:171-190. 177 Libes, M., Boudouresque, C. F. (1987). Uptake and long distance transport of carbon in the marine phanerogam Posidonia oceanica. Mar. Ecol. Prog. Sere 38:177-186.

Lobban, C. S., Harrison, P. J., Duncan, M. J. (1985). The physiological ecology of seaweeds. Cambridge University Press, London.

LUft, J. H. (1971). Ruthenium red and violet I: Chemistry, purification, methods of use for electron microscopy and mechanism of action. Anat. Rec. 171:327-368.

Maki, J. S., Rittschof, D., Samuelsson, M.-a., Szewzyk, U., Yule, A. B., Kjelleberg, S., Costlow, J. D., Mitchell, R. (1990). Effect of marine bacteria and their exopolymers on the attachment of barnacle cypris larvae. Bull. Mar. Sci. 46:499-511.

Marshall, K. C., stout, R., Mitchell, R. (1971a). Mechanisms of the initial events in the sorption of marine bacteria to surfaces. J. Gen. Microbiol. 68:337-348.

Marshall, K. C., stout, R. Mitchell, R. (1971b). Selective sorption of bacteria from seawater. Can. J. Microbiology 17:1413-1416.

Mazzella, L., Cinelli, F., ott, J. A., Klepal, W. (1981). ,Studi sperimentali "in situ" sull' epifitismo della Posidonia oceanica Delile. Quad. Lab. Tecnol. Pesca 3:481-492.

McCarthy, C.A., Pell, B.K., Holburn, C.M., Moore, S.R., Perry, Goddard, D.H., Kirk, A.P. (1979). A tannic acid based preparation procedure which enables leucocytes to be examined SUbsequently by either SEM or TEM. J. Microsc. 137(1):57-64.

McComb, A. J., Cambridge, M. L., Kirkman, H., KUo, J. (1981). The biology of Australian seagrasses. In Pate, J. S., McComb, A. J. (eds.) Biology of Australian plants. University of Western Australia Press, Perth.

McDermid, K. J. (1988). community ecology of some intertidal SUbtropical algae, and the biology and taxonomy of Hawaiian Laurencia (Rhodophyta). Ph.D. dissertation, University of Hawaii. Honolulu, Hawaii. McRoy, c. P. (1974). Seagrass productivity: carbon uptake experiments in eelgrass, Zostera marina. Aquaculture 4:131-137.

178 McRoy, C. P., Barsdate, R. J. (1970). Phosphate absorption in eelgrass. Limnol. Oceanogr. 15:6-13. McRoy, C. P., Barsdate, R. J., Nebert, M. (1972). Phosphorous cycling in an eelgrass (Zostera marina) ecosystem. Limnol. Oceanogr. 17:58-67. McRoy, C. P., Goering, J. J. (1974). Nutrient transfer between the seagrass Zostera marina and its epiphytes. Nature 248:173-174. McRoy, C. P., Goering, J. J., Chaney, B. (1973). Nitrogen fixation associated with seagrasses. Limnol. Oceanogr. 18:998-1002. Mirelman, D., Ofek, I. (1986). Introduction to microbial lectins and agglutinins. In: Mirelman, D. (ed.) Microbial lectins and agglutinins. John Wiley & Sons, Inc., New York. Moriarty, D. J. W., Hayward, A. C. (1982). Ultrastructure of bacteria and proportion of Gram-negative bacteria in marine sediments. Microb. Ecol. 8:1-14. Moriarty, D. J. W., Iverson, R. (1985). Exudation of organic carbon by the seagrass Halodule wrightii and its effect on bacteria growth in the sediment. J. Exp. Mar. BioI. Ecol. 96:115-126. Murray, L., Wetzel, R. L. (1987). Oxygen production and consumption associated with the major autotrophic components in two temperate seagrass communities. Mar. Ecol. Prog. Sere 38:231-239. Novak, R. (1984). A study in ultra-ecology: microorganisms on the seagrass Posidonia oceanica. Mar. Ecol. 5:143-190. Palmer, M. A. (1988). Dispersal of marine meiofauna: A review and conceptual model explaining passive transport and active emergence with implications for recruitment. Mar. Ecol. Prog. Sere 48:81-91. Patriquin, D.G., Knowles, R. (1972). Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. BioI. 16:49 58. Penhale, P. A. (1977). Macrophyte-epiphyte biomass and productivity in an eelgrass (Zostera marina L.) community. J. Exp. Mar. BioI. Ecol. 26:211-224.

179 Penhale, P. A., Smith Jr., W. O. (1977). Excretion of dissolved organic carbon by eelgrass (Zostera marina) and its epiphytes. Limnol. Oceanogr. 22:400-407. Preiss, J. (1989). Chemistry and metabolism of intracellular reserves. In: Poindexter, J. S., Leadbetter, E. R. (eds.) Bacteria in nature vol. 3: structure, physiology, and genetic adaptability. Plenum Press, New York. Rosenberg, M., Kjelleberg, S. (1986). Hydrophobic interactions: role in bacterial adhesion. Adv. Microb. Ecol. 9: 353-393.

Roughgarden, J., Gaines, S. D., Pacala, S. W. (1986). Supply side ecology: the role of physical transport processes. In: Gee, J. H. R., Giller, P. S. (eds.) Organization of communities: past and present. Blackwell Scientific Publications, Oxford. Salisbury, F. B., Ross, C. W. 1992. Plant physiology 4th ed. Wadsworth Pub. Co., Calif. Sand-Jensen, K. (1977). Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot. 3:55-63. Sannes, P. L., Katsuyama, T., Spicer, S. S. (1978). Tannic acid-metal salt sequences for light and electron microscopic localization of complex carbohydrates. J. Histochem. Cytochem. 26:55-61. Santelices, B. (1990). Patterns of reproduction, dispersal and recruitment in seaweeds. Oceanogr. Mar. BioI. Annu. Rev. 28:177-276. Shively, J. M. (1974). Inclusion bodies of prokaryotes. Annu. Rev. Microbiol. 28:167-187. Singley, C. T., Solursh, M. (1980). The use of tannic acid for the ultrastructural visualization of hyaluronic acid. Histochem. 65:93. smith, W.O., Penhale, P. A. (1980). The heterotrophic uptake of dissolved organic carbon in eelgrass (Zostera marina) and its epiphytes. J. Exp. Mar. BioI. Ecol. 48:233-242. Tanaka, N. (1986). Adhesive strength of epiphytic diatoms on various seaweeds. Bul. Jap. Soc. Sci. Fish. 52:817 821.

180 van den Ende, G., Haag, P. (1963). Beobachtungen ueber den epiphytenbewuchs von Zostera marina L. an der bretonischen kueste. Bot. Mar. 5:105-110. van Loosdrecht, M. C. M., Lyklema, J., Norde, W., Zehnder, A. J. B. (1989). Bacterial adhesion: a physicochemical approach. Microb. Ecol. 17:1-15. Waldron, K. W. Baydoun, E. A.-H., Brett, C. T. (1989). Comparison of cell wall composition of tissues from the seagrasses Halophila and Halodule. Aquat. Bot. 35:209-218. Wanner, 0., Gujer, W. (1986). A mUltispecies biofilm model. Biotech. Bioengineer. 28:314-328. Wetzel, R. G. (1983). Attached algal-substrata interactions: fact or myth, and when and how? In: Wetzel, R. G. (ed.) Periphyton of freshwater ecosystems. w. Junk Publishers, The Hague. Wilson, F. S. (1990). Temporal and spatial patterns of settlement: a field study of molluscs in Bogue Sound, North Carolina. J. Exp. Mar. BioI. Ecol. 139:201-220.

181 CHAPTER 4 LECTINS PROBE MOLECULAR FILMS IN BIOFOULING: CHARACTERIZATION OF EARLY FILMS ON NON-LIVING AND LIVING SURFACES

Abstract Films that form quickly on surfaces immersed in the ocean influences the settlement of microbes and eukaryotic settlers via the differential distribution of specific receptors that initiate attachment and metamorphosis. This study probed both inert, non-biological surfaces (glass slides) and living surfaces (leaves of the seagrass, Halophila hawaiiana) with fluorescently labelled lectins to detect and describe distributions of glycoconjugates. These molecules appear to play critical roles as cues to larval settlement. Lectins from Concavalia ensiformis (Con A) and Limulus polyphemus (limulin) bound particular glycoconjugates in one and three-day films that formed following immersion of glass slides in Pearl Harbor. A complex spatial pattern in film development was observed on the glass; approximately circular areas of reduced receptor densities were interspersed in an otherwise homogeneous matrix. The circular area surrounded centrally located aggregates of particles (inclUding microbes) in one-day films. The diameter of these zones increased nearly ten- fold when one and three-day films were compared. A third lectin from Helix pomatia localized a third set of

182 glycoconjugates in specific filmed areas and with the organic matter associated with bacteria on three-day films. Each lectin also bound to surface glycoconjugates on newly emergent leaves of H. hawaiiana with qualitatively different results than those observed for glass. The distributions of receptors for Con A, H. pomatia and limulin lectins were less dense on seagrass surfaces than on filmed glass. Also, receptors for limulin and H. pomatia lectins were associated with the cell walls and peripheral cytoplasm of seagrass cells, not in surface films on leaves. Con A localized glycoconjugates showed only a faint signal on the cell walls of these young leaves. This technique documents previously undescribed spatial and chemical heterogeneities of early, similarly aged surfaces. This understanding allows the definition and localization of possible microscale cues for biofouling. These early steps are expected to be crucial to the distribution of sessile organisms in marine environments.

introduction Lectins, a class of naturally occurring proteins or glycoproteins that bind carbohydrates specifically and noncovalently, have been known for over 100 years. These proteins (or glycoproteins) have now been identified in a variety of biological systems ranging from viruses and bacteria to plants and specific cell types in mammals.

183 Recently, the scientific community has realized that for many lectins, specific binding of particular glycoconjugates, e.g. glycoproteins, glycolipids or polysaccharides deployed on the surfaces of cells, provides a molecular mechanism for recognition and even attachment at the cell-surface (see Sharon and Lis 1989 for recent review). In the marine environment, sugars and adhesion enhancing factors for which a lectin may be specific can be common in sea water as biological exudates (Tosteson and Zaidi 1974, Baier and Weiss 1975, Tosteson et al. 1976, Tosteson and Corpe 1975, Jimenez et al. 1979, Decho 1990). This dissolved mix of substances coats newly immersed surfaces as an abiotic conditioning film that matures to a biofilm and constitutes an important step that precedes settlement by juvenile stages of many marine organisms. Distributions of glycoconjugates on cell surfaces have been demonstrated for species in a variety of algal divisions, the Cyanophyta, Pyrrhophyta, Raphidophyta, Euglenophyta, Chromophyta and Chlorophyta (Vanni et al. 1981, Surek and Sengbusch 1981, Sengbusch et al. 1982, Sengbusch and Muller 1983, Iman et al. 1984, Kaska et al. 1988). In unicellular green algae, glycoconjugates are involved with cell to cell recognition, adhesion, morphogenesis and wall assembly (Iman et al. 1984, Klis et al. 1985, Goodenough et ale 1986, Schlipfenbacher et ale 1986, Samson et al. 1987). In many of these algal studies,

184 the localization of specific glycoconjugates on cell surfaces was detected via light microscopy and the use of a fluorescent chromophore to label the lectin. Emissions from the chromophores have been visualized with epi-fluorescence microscopy and have localized discrete domains of probe binding on algal surfaces (Vanni et al. 1981, Surek and Sengbusch 1981, Sengbusch et al. 1982, Sengbusch and Muller 1983). At least two common planktonic larvae, Janua brasiliensis, a spiroibid polychaete, and Balanus amphorite, a barnacle, differentially settle in response to specific components in biofilms (Kirchman et al. 1982a,b, Maki et ale 1988). Settlement by Janua occurs after the formation of bacterial films and in the presence of glycoproteins or polysaccharides. Janua settlement is blocked by the addition of Concanavalin A (Con A), a lectin specific to glucose and mannose moieties of larger glycoconjugates (Kirchman et al. 1982a). Settlement and adhesion in one algal system is mediated by lectins in a similar manner to known invertebrate larvae and may be involved in settlement of other algae (Maki and Mitchell 1986). The use of glass slides as substrates for settlement is common in biofouling studies. Similarly, seagrass leaves provide numerous replicate natural surfaces in meadows where fouling by microbes, algae and invertebrates can be heavy (Novak 1984). These two surface types differ in surface

185 tension, hydrophobicity, surface texture and chemistry, and can be used to test if glycoconjugates can be visualized via lectin binding. By comparing glass and seagrass leaf surfaces, we can obtain information on elusive spatial features of early film chemistries. To date, no study has examined the formation of molecular or bacterial films on marine surfaces with fluorescently-labelled lectins. This study builds on the observations in the literature that lectin binding sites on surfaces of marine organisms are routinely visualized with epi-fluorescence microscopy and that at least one glycoconjugate directs larval settlement. This technique was developed to facilitate the rapid characterization of surface films generated in laboratory and field settings. This information has great applicability to understanding early, crucial steps in biofouling of man-made and natural surfaces.

Materials and Methods Laboratory films Films made to test the specificity of Con A lectin binding were generated in the laboratory from reagent grade glycoconjugate sugars of Con A. Drops of 2 M mannose and/or glucose were applied to clean glass slides and oven dried for 12 hours at 56 ac.

186 Natural films OD glass Marine films were generated on replicate clean glass slides immersed in seawater following standard methods at the ONR/University of Hawai'i anti-fouling test site, Pearl Harbor, O'ahu, Hawai'i (Rittschoff et ale 1992). Slides were suspended in seawater from racks by monofilament line within 0.25 meter of the surface. Filmed slides were harvested at one and three day intervals during October 1991 and transported to the lab in coplin jars filled with 0.4 ~m filtered seawater or 2% Formalin in filtered seawater.

Natural films OD a living surface The Hawaiian seagrass, Halophila hawaiiana Doty and Stone was collected from discrete patches that were 2 meters deep and 50 meters off shore from Malaekahana, O'ahu, Hawai'i. Newly emergent leaves (between one and three days old) were selected. These leaves do not have the epiphytic colonization that was evident on older leaves. The seagrass leaves were transported to the lab in 0.4 ~m filtered seawater or 2% Formalin in filtered seawater. Samples were probed for lectin binding and examined on the same day of harvest.

Application of lactins Commercially prepared lectins, each conjugated with a

187 specific fluorescent chromophore, were obtained from Sigma Chemical Co. Specific couplings of lectins and chromophores were selected based on expectations of chemical elements in natural films and non-overlapping chromophore emission wavelengths. Lectins of differing sugar specifities were used to detect different sugar residues on the surfaces. The lectins chosen were derived from Helix pomatia (with Texas Red as chromophore abbreviated as "Helix lectin + Texas Red" or "Helix lectin + FITC"), Limulus polyphemus ("limulin + FITC"), Concavalia ensiformis ("Con A lectin + FITC" or "Con A lectin + TRITC") (see The excitation and emission wavelengths per chromophore are given in Table 4.1. The sugar specificities per lectin are given in Table 4.2. To test for the specificity of lectin-glycoconjugate interaction for Con A lectin binding, artificial films were incubated with a hapten to Con A (0.1 M solution of a methyl-mannoside) along with Con A lectin + FITC. One and three day filmed glass slides were incubated

for 30 min. in solutions of 250 ~m/ml lectin + fluorescent chromophore, in EPES buffer, 50 roM, at pH 7.8. MUltiple lectin staining was carried out using the probe with the highest amount of fluorescent molecule to protein first, washed then stained with subsequent lectins (eg. limulin with 5 moles FITCjmole protein, Con A with 0.5 moles TRITC/mole protein). To distinguish prokaryotic cells from

organic matter, DAPI (10 ~g/ml) was added to the final

188

------. --- lectin solution for the last 5 min. of the incubation. Finally, the samples were fixed for 30 sec. in 2% buffered formalin in filtered seawater. Unbound lectin and DAPI were removed by rinsing samples with three changes of distilled water.

Microscopy Films were examined with an Olympus BHS microscope equipped with a BH2-RFC reflected fluorescent attachment. Emission from the 100 W high pressure mercury burner ranged from ca. 280 to 600 nm. Excitation beams of ultraviolet (ca. 300 to 400 nm with 365 nm max), blue (ca. 360 to 500 nm with ca. 475 nm max) or green wavelengths (ca. 450 to 550 nm with ca. 545 nm max) were selected based on the chromophore in use (see Table 4.1). Observations for quality and spatial distribution of emission signals were made via non fluorescent DPlan APO UVPI 20X and 100X lens. Photomicrographs were recorded on Kodak Gold film (ASA 1600) with a Olympus model PM-lOADS automatic photomicrographic system corrected with color reciprocity. Low level autofluorescence by unprobed control samples were visualized for background emission and recorded when possible.

Results single lectin studies in artificial films Artificially generated films of mannose or glucose,

189 demonstrated the expected positive binding by Con A (Table 4.1, Fig. 4.1) while an unstained artificial film under blue excitation showed no sign of positive lectin localization. The use of the Con A + FITC combination resulted in bright emission revealing film crenulations, a probable artifact of the drying routine (Fig.4.1). Emission by the labelled Con A was undetectable following competitive incubation of an artificial film with a glycoconjugate competitor, a-methyl mannoside.

Lectin studies of one and three-day natural films on glass Photographic exposures of unprobed one and three-day old films on glass slides were possible only occasionally because of autofluorescence from microbes or organic matter (Fig. 4.2). Visualization of early biofilm organisms was enhanced with DAPI stain (Fig. 4.3). Autofluorescence, DAPI, lectin responses and film details were consistent in replicates regardless if samples were preserved in Formalin or probed as fresh films (without fixation or drying). As with the artificial film, emission by the Con A + chromophore was undetectable following co-incubation of one and three-day natural films with the sugar competitor, a methyl mannoside. In contrast, lectin probes of one-day natural marine films on glass slides revealed circular patterns in glycoconjugate distributions, specific to Con A (Fig. 4.4)

190 and limulin (Fig. 4.5) binding sites. In each case, glycoconjugates appear to differ in concentration relative to clusters of central, DAPI-stained bacteria or particles (Figs. 4.4, 4.5). The film matrix was rich in Con A receptors (Table 4.2, glycoconjugates with glucose and/or mannose moieties) where the signals gave bright (high concentration) versus dark (low concentration) receptor regions in the film (Fig. 4.4). The highest concentration of binding sites for limulin was in discrete areas near microcolonies (see arrow Fig. 4.5). MUltiple lectin probes with Con A and limulin on the one-day films revealed no greater film complexity. Three-day-old natural films on glass were similar to one day films in that the spatial heterogeneity of the film was still evident (Figs. 4.6, 4.7). The circular zone diameters increase from 4 to 8 ~m to 35 to 85 ~m. An apparent shift in chemistry of the three-day films was shown by the co-occurring limulin and Con A receptor molecules in the matrix of the film. (Figs. 4.6, 4.7). The Con A + TRITC signal was clearest within the dark circular areas (a confirmed by green excitation, not shown). Aggregates of bacteria that were central to concentric bands in the one day films, were not visible in three-day films although scattered bacteria and microcolonies were found (Figs. 4.6 4.8). Extracellular polymeric secretions from bacteria bound Helix lectin + Texas Red (Fig. 4.8, Table 4.2).

191 Lectin studies of a newly emergent living substrate Initially, whole-mounted seagrass leaves were assessed for background autofluorescence. Chloroplasts were the major source of autofluorescence following excitation by blue wavelengths (Fig. 4.10). Both limulin and Helix lectins bound components in newly emergent seagrass cell walls. Distinct nuclei were resolved using DAPI localization (Fig. 4.10). The use of Limulin + FITC revealed a positive response to apparently diffuse receptors densities in periclinal cell walls and a brighter signal from anticlinal walls in the elongate cells over veins (see arrow Fig. 4.11). Anticlinal cell walls and cell matrices demonstrated a positive response to Helix lectin + FITC (Fig. 4.12). Upper cell matrices were sources of the signal as resolved via through-focal examination. This area of epidermal cells is generally rich in chloroplasts, mitochondria and secretory organelles; the probes may have localized glycoprotein-rich synthesis compounds and/or metabolic products. Epifluorescent emission from the young seagrass surface was attributed to specific binding with the cell walls in contrast to the natural films on glass (compare Figs. 4.11, 4.12 versus 4.6). No microbes or surface films were detected on any young leaves in this study.

192

~-----~--- Discussion Epifluorescence microscopy coupled with lectin specificity offers a new way to reliably and specifically probe the spatial relationships of complex molecular distributions on marine surfaces when films are present. Glycoconjugates were clearly present in these early films on inert surfaces in significant concentrations to allow their visualization and localization with lectin probes. Clear but differential binding of selected lectins also revealed a marked degree of spatial differentiation in early films that formed repeatedly and consistently on glass. In marked contrast, no films were detected on young seagrass leaves. Finally, these results show that with maturation of the conditioning film on glass the identity, spatial pattern and numbers of glycoconjugates change and increase respectively. This evidence for a chemical shift in one to three-day films comes from early dominance of Con A receptors to co dominance of Con A and limulin receptors in the film matrix. The mechanisms that formed circular patterns in the films on glass are not known. Two alternative explanations are possible. The circular patterns could result from periodic precipitation of chemicals in a gel (the film). In this case, diffusion from point sources (inorganic, organic or microbes) would occur as in the formation of "Liesegang" bands (Williams 1938a,b, 1939a,b, Crowle et ale 1963). The edge pattern is then set by the limit of diffusion from the

193 source. Alternatively, the circular patterns could result from the metabolism of central organisms as their metabolic requirements "deplete the surrounding region of nutrients" (Wimpenny 1981). Further investigation is needed to differentiate between these, and possibly other mechanisms that might create the film patterns. Solid surfaces immersed in seawater are modified by absorption of glycoproteins (Baier 1973). From this, early films are viewed as both a precursor to and a physical presence between settled organisms and the immersed surface (eg. Schrader et ale 1988). In fact, a variety of glycoprotein and lectin sites may be present on these surfaces and are likely to enhance the adhesion of cells and other particulate matter by acting as recognition sites or settlement cues (eg. Maki and Mitchell 1986, Kirchman et al. 1982a). Biofilms facilitate adhesion by some organisms or metamorphosis of at least one invertebrate (Kirchman et al. 1982b). Bacterial polysaccharides or glycoproteins in a biofilm have been found to form complementary bonds with lectins as settlement cues in Janua (Kirchman et ale 1982a). Bacterial settlement on the alga Chlorella is enhanced by lectin-like receptors on bacterial surfaces (Imam et al. 1984). Specific associations of algal propagules with macrophytes may involve surface recognition mechanisms (Clarke and Knox 1978, Ducker and Knox 1984). Non-random patterns in settlement can result from surface recognition

194 and selection. The heterogeneous overall filmed glass surface may be viewed as a series of microhabitats over the surface. Patterns in settlement generate patterns in community development and so be determinants of community structure and dynamics. The results of this study show that early films on glass are chemically diverse and spatially dynamic within 24 h of immersion. This is in contrast with comparably aged leaves of Halophila hawaiiana which demonstrates that a biological system can have a different time course, chemistry and spatial patterning for film formation over a three-day period. The heterogeneous pattern of film formation on glass did not occur on the young seagrass leaves. Even though mature seagrass leaves (ca. 12 days) from the same population were commonly fouled, not even microbes occurred on these unfouled young leaves. Rather than have lectin-localized glycoconjugates on leaf surfaces, lectin receptors were associated with outer cell walls in young leaves. There was no discernable surface film. The conclusion is that newly emergent leaves are not attractive to colonists at least in part because common lectin binding sites are not present. Glycoconjugates and lectins (as a class of glycoproteins) have many inter- and intracellular functions in plants. Glycoproteins function as recognition molecules and are active in highly specific cell to cell, molecule to

195 molecule and molecule to cell interactions (Oseroff et al. 1973). Glycoproteins in plant cell walls are involved in growth, organization of cell wall deposition, osmotic relations and permeability that leads to concentration of nutrients and toxins at the cell wall/plasmalemma boundary (Cook and Stoddart 1973). Plant lectins interact with microorganisms in the formation of symbiotic relationships (Etzler 1986, Pusztai 1991), as a means of defence against bacteria and viruses (Etzler 1986, Sharon 1986), and as a source of bacterial infection. As carbohydrate receptor molecules they aid in transport of carbohydrates and membrane stabilization (Etzler 1986). Cellular expression of certain glycoconjugates depends on the developmental and physiological states as illustrated in algal systems (Surek and Sengbusch 1981, Sengbusch et al. 1982, Sengbusch et al. 1985). Changes in the surface carbohydrates lead to differences in their responses to lectins (Pusztai 1991). The information obtained by lectin probes of the young seagrass leaves is a part of the total chemistry of the organism. The interior of a seagrass leaf is a mosaic of chemistries with limulin receptors concentrated in the anticlinal walls and in cells over veins. Helix lectin receptors are found in the anticlinal walls, the organelle rich outer cytoplasm and perhaps the periclinal wall.

196 Seagrass walls are clearly rich in carbohydrates that provide general cell wall functions including structural integrity and, combined with the plasmalemma, regulation and exchange of solutes between the cytoplasm and external environment (Kuo et ale 1990, Waldron 1990). Surface structure components that include Con A receptors such as D glucose and D-mannose on the surface of a unicellular alga have been implicated in a surface recognition process (Reisser et ale 1982). Cell to cell and cell to substratum adhesion are processes whereby cells recognize and adhere to each other in crucial and complicated events. We have localized molecules with known receptor components for both the glass and seagrass surfaces. Future studies are needed to 1) characterize specific receptor molecules, 2) continue to assess the relationships of glycoconjugates with cells, and 3) assess the biological and ecological significances of the molecular mosaic of a surface, relative to surface interactions with propagules.

conclusions Understanding the early chemistry of surfaces on which communities develop may provide means to selectively disrupt the establishment of fOUling communities. Early films that form on inert glass surfaces in the marine environment are chemically complex and spatially heterogeneous. No acellular film was found on young seagrass leaves. This

197 absence was correlated with a lack of microbes on these surfaces. Epifluorescence microscopy allows rapid sensitive probing of surface features. This technique can be used to determine the chemical nature of the film relevant to settlement and the spatial patterns of potential binding sites for settling stages of larvae and algae.

198

------Table 4.1. Excitation and emission wavelengths for fluorescent chromophores available with many lectins.

Chromophore Excitation A Range Emission Max nm nm

FITC UV to 495 (blue) 535 (green) (fluorescein isothiocyanate)

TRITC UV to 557 (green) 576 (yeljoran) (tetramethylrhodamine isothiocyanate)

Texas Red UV to 596 (yeljoran) 615 (oranjred) (sulforhodamine 101 acid chloride)

DAPI 360 (UV) 490 (blue) 4,6-diamidino-2-phenylindole

199 Table 4.2. Sugar specificity and molecular characterizations of common lectins. 1 = Reeke et al. 1974; 2 = Hammerstrom et al. 1972; 3 = Fernandez-Moran et al. 1968; 4 = Marchelonis and Edelman 1968; 5 = Gilbride and Pistole 1979.

Lectin Source Mol. wt. Sugar Specificity Ref. x 10-3

Concanavalin A 102 a-D-man, a-D-glc 1 Canavalia ensiformis

Helix pomatia 79 D-galNAc 2

Limulus polyphemus 400 NeuNAc, D-galNAc 3-5 D-glcNAc

200 Figure 4.1. A positive signal from Con A + FITC bound to a glucose film. Blue wavelength excitation. Scale = 50 ~m for Figures 4.1-4.6, 4.9-4.12.

Figure 4.2. Autofluorescence from unstained three-day films on glass. UV wavelength excitation.

Figure 4.3. Microorganisms in a three-day film on glass localized by DAPI staining. UV wavelength excitation.

Figure 4.4. Distinct spatial pattern of glycoconjugates distribution in the matrix of a one day film on glass revealed by Con A + FITC; OAPI localized bacteria.

Figure 4.5. Lectin probe, limulin + FITC. A circular pattern of distribution for a limulin + FITC specific glycoconjugate (arrow) and OAPI localized microcolony of a one day film on glass. UV wavelength excitation.

Figure 4.6. Complex spatial pattern on glass as revealed by in a three day film on glass was revealed by limulin + FITC, Con A lectin + TRITe, and OAPI staining. Microcolonies overlay glycoconjugate matrix. UV wavelength excitation.

Figure 4.7. Magnified area of Fig. 6 reveals heterogeneity of receptor densities. UV wavelength excitation. Scale = 1.5 ~m for Figures 4.7-4.8.

Figure 4.8. Glycoconjugates specific to Helix lectin (+ Texas Red) are localized in the extracellular polymeric secretions (arrow) of DAPI stained bacteria in three day films on glass. UV wavelength excitation.

Figure 4.9. Autofluorescence by cells of newly emergent leaves of Halophila hawaiiana. Blue wavelength excitation.

Figure 4.10. DAPI stained nuclei of seagrass cells. UV wavelength excitation.

Figure 4.11. Epidermal cells as revealed via limulin (+ FITC). Lectin receptors were concentrated in anticlinal walls over veins (arrow). Blue wavelength excitation. Figure 4.12. Epidermal cell walls as revealed via Helix lectin + FITC. Lectin receptors were concentrated in the upper cell matrix and anticlinal walls. Blue wavelength excitation.

201 202 Literature cited Baier R. (1973). Influence of initial surface conditioning of materials on bio-adhesion. In: Proc. 3rd Int. congo on marine corrosion and fouling. Northwestern University Press, Illinois. Baier, R. E., Weiss, L. (1975). Demonstration of the involvement of adsorbed proteins in cell adhesion and cell growth. Adv. Chem. Sere 145:300-307. Characklis, W. G., Marshall, K. C. (1990). Biofilms: a basis for an interdisciplinary approach. In: Characklis, w. G., Marshall, K.C. (eds.) Biofilms. John Wiley & Sons, Inc., New York. Clarke, A. E., Knox, R. B. (1978). Cell recognitions in flowering plants.- Quart. Rev. BioI. 53:3-28. Cook, G. M. W., Stoddart, R. W. (1973). Surface carbohydrates of the eukaryotic cell. Academic Press, London. Crowle, A. J., Lueker, D. C., Gaskill, H. S. Jr. (1963). Periodic (Liesegang) precipitation of chemicals. Nature 199:623-624. Ducker, S. C., Knox, R. B. (1984). Epiphytism at the cellular level with special reference to algal epiphytes. In: Linskens, H.F., Heslop-Harrison, J. (eds.) Cellular interactions. Springer-Verlag, Berlin. Decho, A. W. (1990). Microbial exopolymer secretions in ocean environments: Their role(s) in food webs and marine processes. Annu. Rev. Oceanogr. Mar. BioI. 28:73-153. Fernandez-Moran, H., Marchalonis, J. J., Edelman, G. M. (1968). Electron microscopy of a hemagglutinin from Limulus polyphemus. J. Mol. BioI. 32:467-469. Gilbride, K. J., Pistole, T. G. (1979). Isolation and characterization of a bacterial agglutinin in the serum of Limulus polyphemus. Prog. Clin. BioI. Res. 29:525-535. Goldstein, I. J. (1975). In: Chowdhury, T. K., Weiss, A. K. (eds.) Concanavalin A. Plenum, New York. Goodenough, U. W., Gebhart, B., Mechan, R. P., Heuser, J. E. (1986). Crystal of Chlamydomonas reinhardii cell wall: pOlymerization, depolymerization and purification of glycoprotein monomers. J. Cell BioI. 103:405-418.

203

---_. ------Hammarstrom, S., Weston, A., Bjork, I. (1972). Subunit structure of Helix pomatia A hemagglutinin. Scand. J. Immunol. 1:295-309. Imam, S. H., Bard, R. F., Tosteson, T. R. (1984). Specificity in marine microbial surface interactions. App. Envir. Micro. 48:833-839. Ivatt, R. J. (1984). The biology of glycoproteins. Plenum Press, New York. Jimenez, B. R., Tosteson, T. R., Zaidi, B. R., Tasi, T. S., Atwood, D. K. (1979). Adhesion enhancing organics in the Eastern Caribbean Sea. In: Stewart, H.B. (ed). CICAR II symposium of Progress in Marine Research in the Caribbean and Adjacent Regions. FAO Fisheries Report 2000. Rome. Kaska, D. D., Polne-Fuller, M., Gibor, A. (1988). Developmental studies in Porphyra (Bangiophyceae). II. Characterization of the major lectin binding glycoproteins in differentiated regions of the Porphyra perforata thallus. J. Phycol. 24:102-107. Kirchman, D., Graham, S., Reish, D., Mitchell, R. (1982a). Lectins may mediate in the settlement and metamorphosis of Janua (Dexiospira) brasiliensis Grube (Polychaeta: Spirorbidae). Mar. BioI. Lett. 3:131-142. Kirchman, D., Graham, S., Reish, D., Mitchell, R. (1982b). Bacteria induce settlement and metamorphosis of Janua (Dexiospira) brasiliensis Grube (Polychaeta: Spirorbidae). Exp. Mar. BioI. Ecol. 56:153-163. Klis, F. M., Samson, M. R., Touw, E., Musgrave, A., van den Ende, H. (1985). Sexual agglutination in the unicellular green algal Chlamydomonas eugametos. Identification and properties of the mating type plUS agglutination factor. Plant Physiol. 159:740-746. Kornfeld, R., Kornfeld, S. (1970). The structure of the phytohemagglutinin receptor site from human erythrocytes. J. BioI. Chem. 245:2536-2545. KUo, J., Ridge, R. W., Lewis, w. V. (1990). The leaf internal morphology and ultrastructure of Zostera muelleri Irmisch ex. Aschers. (Zosteraceae): a comparative study of the intertidal and subtidal forms. Aquat. Bot. 36:217-236.

204 Maki, J. S., Rittschof, D., Costlow, J. D., Mitchell, R. (1988). Inhibition of attachment of larval barnacles, Balanus amphorite, by bacterial surface films. Mar. BioI. 97:199-206. Maki, J. S., Mitchell, R. (1986). The function of lectins in interactions among marine bacteria, invertebrates, and algae. In: Mirelman, D. (ed.) Microbial lectins and agglutins. John Wiley and Sons, New York. Marchelonis, J. J., Edelman, G. M. (1968). Isolation and characterization of a hemagglutinin from Limulus polyphemus. J. Mol. BioI. 32:453-465. Novak, R. (1984). A study in ultra-ecology: microorganisms on the seagrass Posidonia oceanica. Mar. Ecol. 5:143-190. Pennell, R. I., Knox, J. P., Scofield, G. N., Selvengran, R. R., Roberts, K. (1989). A family of abundant plasma membrane-associated glycoproteins related to the arabinogalactan proteins in unique to flowering plants. J. Cell BioI. 108:1967-1977. Pusztai, A. (1991). Plant lectins. Cambridge University Press, Cambridge. Odham, G., Tunlid, A., Valeur, A., Sundin, P., White, D. C. (1986). Model system for studies of microbial dynamics at exuding surfaces such as the rhizosphere. Appl. Envir. Microbiol.52(1):191-196. Oseroff, A. R., Robbins, P. W., Burger, M. M. (1973). The cell surface membrane: biochemical aspects and biophysical probes. A. Rev. Biochem. 42:647-682. Rittschoff, D., Claire, A. S., Gerhardt, D. J., Bonaventura, J., Smith, C. M., Hadfield, M. G. (1992). Short term field assessment of antifouling and foul-release coatings. Biofouling, in press. Reek, G. N., Becker, J. W., Cunningham, B. A., Gunter, G. R., Wang, J. L., Edelman, G. M. (1974). Relationships between the structure and activities of Concanavalin A. Ann. N.Y. Acad. Sci. 234:369-382. Reisser, W., Radunz, A., wiessner, W. (1982). participation of algal surface structures in the cell recognition process during infection of aposymbiotic Paramecium bursaria with Symbiotic chlorellae. Cytobios 33:39-50.

2~ Samson, M. R., Klis, F. M., Homan, H. L., van Egmond, P., Musgrave, A., van den Ende, H. (1987). composition and properties of the sexual agglutinins of the flagellated green alga Chlamydomonas eugametos. Planta 170:314-22. Schlipfenbacher, R., Wenzel, S., Lottspeich, F., Sumper, M. (1986). An extremely hydroxyproline-rich glycoprotein is expressed in inverting Volvox embryos. FEBS Lett. 209:57-62. Schrader, M. E., Cardamone, J. A., Loeb, G. I., Arora, O. P. (1988). Electron spectroscopy and conditioning films on metal. In: Thompson, M. F., Sarojini F. R., Nagabhushanam, R. (eds.) Marine biodeterioration: advanced techniques applicable to the Indian Ocean. A.A. Balkema, Rotterdam. Sengbusch, P. v., Mix, M., Wachholz, I., Manshard, E. (1982). FITC-Iabelled lectins and Calcofluor White ST as probes for the investigation of the molecular architecture of cell surfaces. Studies on Conjugatophycean species. Protoplasma 111:38-52.

Sengbusch, P. v., MUller, U. (1983). Distribution of glycoconjugates at algal cell surfaces as monitored by FITC-conjugated lectins. Studies 0n selected species from Cyanophyta, Pyrrhophyta, Raphidophyta, Euglenophyta, Chromophyta and Chlorophyta. Protoplasma 114:103-113.

Sharon, N., Lis, H. (1989). Lectins as cell recognition molecules. Science 246:227-234.

Tosteson, T. R., Atwood, D. K., Tsai, R. S. C. (1976). Surface active organics in the Caribbean Sea. In: Oceans 76 symposium. Marine Technology Society. New York.

Tosteson, T. R., Corpe, W. A. (1975). Enhancement of adhesion of marine Chlorella vulgaris to glass surface. Can. J. Microbiol. 21:1025-1031.

206 Tosteson, T. R., Revuelta, R., Zaidi, B. R., Imam, S. H., Bard, R. F. (1985). Aggregation-adhesion enhancing macromolecules and the specificity of marine microbial surface interactions. J. Colloid and Interface Science 104(1):60-71. Vanni, G. L., Bonora, A., Dall'olio, G. (1981). Distribution of the receptors for Concanavalin A on the surface of Euglena gracilis as revealed by fluorescence microscopy. Plant Sci. Lett. 22:23-28. Waldron, K. W., Baydoun, E. A., Baydoun, H., Brett, C. T. (1989). Comparison of cell wall composition of tissues from the seagrasses Halophila and Halodule. Aquat. Bot. 35:209-218. Williams, J. W. (1938a). Bacterial growth 'spectrum' analysis I. Methods and application. Am. J. Med. Tech. 4:58-61. Williams, J. W. (1938b). Bacterial growth 'spectrums'. II. Their significance in pathology and bacteriology. Am. J. Path. 14:642-645. Williams, J. W. S. (1939a). The nature of gel mediums as determined by various gas tensions and its importance in growth of microorganisms and cellular metabolism. Growth 3:181-196. Williams, J. W. S. (1939b). Growth of microorganisms in shake cultures under increased oxygen and carbon dioxide tensions. Growth 3:21-33. Wimpenny, J. W. T. 1981. Spatial order in microbial ecosystems. BioI. Rev. 56:295-324.

207 CHAPTER 5

SYNTHES:IS The growth characteristics of seagrasses as well as their population architecture and large replicate surface area contribute to a model system for the study of colonization in the marine environment. The interactions, generated as epiphytes recruit to seagrass leaves, may identify factors that structure community development in benthic marine environments. The importance of the substrate in colonization was considered in the literature review. The living substrate, Halophila hawaiiana, is introduced in Chapter 1 from taxonomic and ecological perspectives. The ubiquity of seagrass epiphyte communities are acknowledged and their function at the ecosystem level is summarized from other systems. Patterns in colonization have been observed in previous studies and the process of colonization is further considered. Molecular-level forces and fine-scale surface properties may explain patterns in some systems. The Hawaiian seagrass is the base of a productive meadow community. The anatomy of mature leaves is reported in Chapter 2. The objective of this chapter was to document the ultrastructure of the seagrass leaves as a basis for further probes of fine-level species interactions in the epiphyte communities. The ultrastructure of the surface tissue was contrasted in leaves from wave exposed and

208 sheltered environments. The leaves were found to have similar and consistent epidermal cell ultrastructure from both sites. This chapter is the first documentation of the fine structure of this species. Many observations in this study on epidermal, ground and vascular tissues extend our knowledge of anatomy for the genus. Epiphytized leaves from the wave exposed and sheltered sites were then assessed for ultrastructure of seagrass and epiphytes in Chapter 3. The research was organized around the null hypothesis, wave exposure does not influence colonization of seagrass leaves by epiphytes. The sequential null hypotheses are further considered: 1) there are no differences in the affects of colonists on host leaves at these two sites; 2) there are no differences in the specific affects of different taxa of colonists on the host, and 3) there are no differences in subsequent colonization depending on taxa of colonists present. Each hypothesis was rejected in that: 1) the epiphytes differed at each site, particularly in respect to the taxa that adhered directly to the seagrass leaves; 2) some taxa of epiphytes that were directly attached to the leaves were associated with epidermal cell modifications; 3) the form of the epiphyte communities differed in each site indicating the primary colonists affected subsequent colonization. In the wave exposed environment, leaves were a spatial substrate for colonists; crustose corallines settled on

209 leaves and were associated with anatomical alterations including elaborate cell wall ingrowths but no reduction in numbers of chloroplasts and mitochondria. The community may be characterized by spatial and possibly physiological interdependence between overall primary productivity is maintained. In the wave sheltered environment, the leaves appeared to provide a physiological substrate for colonists~ bacteria settled on leaves and were associated with anatomical alterations including cell wall lysis, alteration of the cell wall ingrowth region and reduced numbers of chloroplasts and mitochondria. The community may be characterized by physiological dependence of the epiphytes on the host leading to loss of photosynthetic function and the demise of the host leaves. Chapter 3 also contributes information on potential surface active molecules at the seagrass and epiphyte interface~ distinctive polysaccharides were envisioned at cell surfaces. In other systems these molecules are a specific source of adhesives in cell attachments. The Objective of Chapter 4 was to address a potential basis for differential recruitment as occurred in the wave exposed and sheltered sites. The previous 2 chapters addressed aspects of the mature leaves and leaf communities albeit only 12 days old. The SUbsequent chapter assessed a specific class of surface molecules that have been

210 considered important in prebiotic "conditioning" of surface marine surfaces and in cell to cell interactions. Distributions and identities of naturally occurring glycoconjugates were probed by the novel use of specifically labeled lectins. Incipient stages of colonization could be contrasted for different substrates by this technique. The results clearly indicate that the surfaces differed in respect to abiotic and biotic film formation. The film that formed between one and three days on glass slides was rich in glycoconjugates. No abiotic film or microbes were visible on three-day old seagrass leaves. Glycoconjugates were visible in specific areas of the leaf's interior. Colonization may be driven by specific lock and key mechanisms where glycoconjugates provide one half of specificity and settling stages provide the other half. The genesis of patterns in community development may involve these earliest events in colonization.

211

------_._---