<%BANNER%>

Functional Ecology of the Gametophytes and Sporophytes of Tropical Ferns


PAGE 1

FUNCTIONAL ECOLOGY OF THE GAMETOPHYTES AND SPOROPHTYES OF TROPICAL FERNS By JAMES EDWARD WATKINS, JR. A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

To my parents James E. Watkins and Juanita P. Watkins for encouraging my curiosity in the natural world, for spending hours looki ng for material to add to my annual elementary-school leaf collections, and for wr iting sick notes to my high school principal so that I could skip class and spend the early days of Spring in search of the elusive Botrychium lunarioides

PAGE 3

iii ACKNOWLEDGMENTS Many people and organizations contributed to the successful completion of my work. I thank my supervisory committee me mbers (Stephen Mulkey, Michelle Mack, Thomas Sinclair, and Pamela Soltis) for their many contributions and their patience throughout. I also thank members of th e University of Florida Ecology Group (specifically Louis Santiago, Juan Posada, Grace Crummer, Jordan Major and Jason Vogel) for all of their help. I am especial ly indebted to Jason Vogel for his countless hours of statistical discussions and for his abil ity to raze my radical political ambitions. Throughout my doctoral studies, I often relied on the sage advi ce of Dr. Jack Ewel be it academic, personal, or hunting: I am gratef ul. I also thank Robbin Moran (New York Botanical Garden), and Donald Farrar (Iowa State University, Ames) for always lending an ear or helping hand when it was needed most. Some of this work took place at La Selva Bi ological Station in Costa Rica and I am indebted to the Organization for Tropical Studies for opening the doors to the tropical world to me. I also thank my wife, Cather ine Cardels, for her immense patience and support throughout this process. I also thank my son Santiago for teaching me what life is really all about. Funding was provide d by the National Science Foundation, the Organization for Tropical Studies, the Mell on Foundation, the American Fern Society, and the University of Florida Gradua te School and Department of Botany.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 2 GAMETOPHYTE ECOLOGY AND DEMOGRAPHY OF TROPICAL EPIPHYTIC AND TERRESTRIAL FERNS...............................................................5 Introduction................................................................................................................... 5 Materials and Methods.................................................................................................6 Study Site...............................................................................................................6 Gametophyte Transects.........................................................................................7 Disturbance Plots...................................................................................................7 Demography..........................................................................................................8 Gametophyte Survival Analysis............................................................................9 Results........................................................................................................................ .10 Transects..............................................................................................................10 Disturbance Plots.................................................................................................11 Demography........................................................................................................12 Discussion...................................................................................................................14 Gametophyte Distributions..................................................................................14 Density and Species Richness.............................................................................17 Sporophyte Ecology............................................................................................18 Conclusions.................................................................................................................19 3 COMPARATIVE DESICCATION TO LERANCE OF TROPICAL FERN GAMETOPHYTES: ECOLOGICAL AND EVOLUTIONARY CONSEQUENCES.....................................................................................................29 Introduction.................................................................................................................29 Materials and Methods...............................................................................................33

PAGE 5

v Spore Material and Growth Conditions...............................................................33 Desiccation Experiments.....................................................................................33 Chlorophyll-Fluorescence Measurements...........................................................35 Statistical Analysis..............................................................................................35 Results........................................................................................................................ .36 Desiccation Survey..............................................................................................36 Desiccation Rates................................................................................................37 Desiccation Cycles..............................................................................................38 Discussion...................................................................................................................38 Variation in VPD.................................................................................................39 Desiccation Cycles..............................................................................................41 Conclusions.................................................................................................................43 4 NITROGEN-15 NATURAL ABUN DANCE AND NITROGEN USE STRATEGIES OF THE GAMETOP HYTES AND SPOROPHYTES OF TROPICAL EPIPHYTIC AN D TERRESTRIAL FERNS.........................................55 Introduction.................................................................................................................55 Material and Methods.................................................................................................57 Study Site.............................................................................................................57 Study Species.......................................................................................................57 Isotopic Natural Abundance and 15N Labeled Uptake......................................59 Nutrient Uptake Calculations..............................................................................60 Results........................................................................................................................ .60 15N Natural Abundance and N concentration (mg g-1)......................................60 15N Labeled Uptake...........................................................................................61 Discussion...................................................................................................................62 Conclusions.................................................................................................................67 5 CONCLUSIONS........................................................................................................75 LIST OF REFERENCES...................................................................................................79 BIOGRAPHICAL SKETCH.............................................................................................88

PAGE 6

vi LIST OF TABLES Table page 2-1 Relationship of gametophyte density and richness with three levels of experimental disturbance and two light levels.........................................................21 2-2 Demographic and survival analyses for the gametophytes of 5 fern species using the Wilcoxon test to compare survival dist ribution functions for different species.22 3-1 Species and life form from the initial desiccation survey........................................44 3-2 Fv/Fm recovery results from the re peated measures ANOVA for gametophytes exposed to three different desiccation intensities.....................................................45 3-3 Fv/Fm recovery results from the re peated measures ANOVA for gametophytes exposed to 1, 2, or 3 desiccation cycles...................................................................46 4-1 Species, life form and ecology for the natural abundance and uptake experiments68

PAGE 7

vii LIST OF FIGURES Figure page 2-1 Number of gametophytes counted and thei r relation to disturbance from natural transects....................................................................................................................23 2-2 The percentage of fern gametophytes as influenced by type of disturbance............24 2-3 The relationship between canopy openness and gametophyte density for A. canopy and B. terrestrial species..............................................................................25 2-4 Gametophyte densities as influenced by light and disturbance in experimental plots.......................................................................................................................... 26 2-5 Mean longevity (months) A) and pe rcent gametophytes st ill alive and unrecruited B) for the 25-m onth period of the study...................................................27 2-6 Kaplan-Meier survivorship curves A) and proportion recruiting B) of 5 species of fern gametophytes over the 25-mont h study period. No data were collected during months 4-7 and 15-24...................................................................................28 3-1 A) Gametophyte drying curves from 12 tropical fern species of different habitats. Species were exposed to a VPD to 1.32KPa (~50%RH) for 45 min. B) Depression of photochemical efficiency in the same gametophytes over a series of decreasing thallus water contents.........................................................................48 3-2 The rate of absolute water loss rela tive to gametophyte size as indexed by dry mass..........................................................................................................................4 9 3-3 A) Rate of thallus drying as calculate d from 3-1 for 12 tropical fern species of different habitats. (B) Proportional recove ry of the pre-treatment dark adapted value of Fv/Fm in these same species......................................................................50 3-4 Proportional recovery of the pretreatment dark adapted value of Fv/Fm and rate of thallus water loss expressed as A) re lative water content ((g fresh weightg dry weight)/g saturated weight g dr y weight))*100 and B) absolute water content (g wet wei ght/g dry weight)........................................................................51

PAGE 8

viii 3-5 Fv/Fm recovery graphs for gametophytes of A) Diplazium subsilvaticum, B) Phlebodium pseudoaureum (c) Polypodium triserale exposed to three different desiccation intensities: VPD~0.53kPa (20%RH), VPD~1.32kPa (50%RH), and VPD ~2.12kPa (80%RH).........................................................................................52 3-6 Proportional Fv/Fm recovery results for gametophytes exposed to 1, 2, or 3 desiccation cycles at VPD~ 1.32kPa (50%RH).......................................................53 3-7 Morphology in fern gametophytes is di verse and is closely related to species ecology and phylogeny.............................................................................................54 4-1 Sporophtyic and Gametophytic 15N natural abundance sign atures of 10 tropical fern species...............................................................................................................69 4-2 Sporophtyic and Gametophytic (a) 15N natural abundance signatures and (b) N concentration (mg g-1) of epiphytic, terrestrial, an d hemiepiphytic tropical fern species......................................................................................................................70 4-3 15N natural abundance signatures from the hemiepiphytic fern Lomariopsis vestita .......................................................................................................................71 4-4 Uptake curves from 15N labeled solutions..............................................................72 4-5 Uptake saturation values (Vmax) of each N form derived from Michaelis-Menten functions of the data from Fig. 4-2...........................................................................73 4-6 Uptake saturation values (Km) of each N form derived from MichaelisMenten functions of the data from Fig. 4-2.............................................................74

PAGE 9

ix Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FUNCTIONAL ECOLOGY OF THE GAM ETOPHYTES AND SPOROPHTYES OF TROPICAL FERNS By James E. Watkins, Jr. May 2006 Chair: Stephen S. Mulkey Cochair: Michelle Mack Major Department: Botany Ferns are an important part of both temp erate and terrestrial floras, yet their ecology remains poorly understood. Although fe rns are dispersed by tiny wind-blown spores, most species are limited to specific habitats; on local leve ls, ferns are no more widespread than angiosperms. One aspect of fern biology that poses unique ecological problems is the dependence on a free-living gametophyte. I examined the autecology and ecophysiology of the fern gametophyte to understa nd this structures role in shaping fern distributions. My study showed that the game tophytes of epiphytic and terrestrial ferns respond differently to light, disturbance, and desiccation stress, and show unexpected versatility in nutrient relations In all cases, such variation is closely linked to species ecology. Selective pressures acting on th e gametophyte generation may be largely responsible for species distributions.

PAGE 10

1 CHAPTER 1 INTRODUCTION The ferns, with some 12,000 species, are the third most species group of land plants following angiosperms and bryophytes. Their intermediate evolutionary position affords the group a combination of both non-vasc ular and vascular pl ant life histories. Ferns rely on a supposedly delicate, shor t-lived, and usually haploid independent gametophyte (one of their connections to non-va scular plants), and in some, recruitment from the gametophyte often follows into the usually diploid sporophyte stage with lignified vascular tissue (their most obvious connection to va scular plants). It is the ecology of the gametophyte in this choreographe d alternation of generations that remains largely unstudied. Such scientific deficit has been commented upon for decades (Pickett, 1914; Holttum, 1938; Cousens, Lacey, and Kelly, 1985; Greer and McCarthy, 1999), yet there has been little movement to increase our understanding of basic fern ecology. Richard Eric Holttum (1895-1990) is arguabl y the founding father of fern ecology. Professor Holttum was trained at Cambridge and spent much of his time teaching and studying the ferns of Southeast Asia. By his ha nd one of the first great treatises of fern ecology was written (Holttum 1938). His seminal The ecology of tropical pteridophytes first approached the topic in a synthetic way by incorporating intimate knowledge of ferns and combining it with careful observat ion and questioning to get to a big-picture conclusion about the ecology of the group. He also addressed ecology of the gametophyte generation:

PAGE 11

2 We are accustomed to see and to marvel at the great varied form and adaptation of the sporophtyes, which are the ferns as we know them, but indeed there must be nearly as much variety of adaptation among the gametophytes. It is true that if the prothallus of Platycerium grew upon the forest floor, the resulting sporophyte, if produced, would find itself in uncongeni al surroundings, and would not develop very far; but it is also true that the Platycerium prothallus must be able to develop in relatively exposed position on the tree trunk in which prothalli of many ferns would be unable to exist (Holttum 1938, pp. 421-422). One of Holttums key observations was the recognition of gametophyte-mediated controls on fern recruitment. Platycerium commonly know as staghorn ferns, are Old World epiphytes with diverse ecology; many species grow on highly exposed emergent canopy tree trunks, but never on the forest fl oor. This observation combined with the copious production of wind-dispersed spores showed that dispersal is perhaps of only modest importance in the distri bution of ferns: the gametophyte played the critical role in recruitment. Unfortunately, Holttums call to arms was largely ignored and we have progressed little since th e publication of his work. Many factors have limited the study of fern gametophyte ecology. Avoidance of the gametophyte generation may have been driven by some of the comments made by Frederick Orpen Bower (Bower, 1923) in The Ferns. Bower saw little taxonomic value in fern gametophytes and doubted their utility in advancing pteridology. Bower was such a recognized figure (and the de pauperate literature of th e time basically supported his ideas) that many botanists took his words to heart. Fortunately, many studies have since incorporated gametophytic characters into phylogenies, and we have learned that gametophytes have systematic value and have many taxonomic characters that allow for species or morpho-type identification (A tkinson and Stokey, 1964; Nayar and Kaur, 1971; Chiou and Farrar, 1997; Watk ins Jr. and Farrar, 2005).

PAGE 12

3 There has been limited application of modern ecological methodology and statistics to the study of fern gametophyte eco logy. In a series of observational field and elegant lab experiments, Pickett (1913; 1914) showed that the gametophytes of Asplenium rhizophyllus and A. platyneuron could be long-lived and survive winter temperatures and extreme drought Picketts wo rk helped usher in a new way of thinking about fern gametophyte ecology and later re ports by Mottier (1927) and Walp (1951) showed that the gametophytes of some temper ate species were esse ntially indeterminate and could grow in vitro for decades, if reproduction were prevented. This notion was further supported by Donald Farrar who de veloped the now classic story of tropical gametophytes growing independently of s porophytes in the temperate Appalachian Mountains (Farrar, 1967, 1971; Farrar, 1998) In many cases, these species form populations of asexually reproduc ing gametophytes that number in the tens of thousands and appear to survive winter freezes and summer droughts. In a series of papers, Michael Cousens de veloped the concept of gametophyte safe sites and showed that multiple factors act at the level of the gametophyte to shape their distribution and recruitment (Cousens, 1979, 1981; Cousens, Lacey, and Kelly, 1985; Cousens, 1988; Cousens, Lacey, and Scheller, 1988). Cousens work largely developed in the backdrop of earlier studies showing that fern gametophytes have a distinct ecology relative to sporophytes. From the turn of the century with Picket ts work, to more recent work by Cousens and Farrar, we have learned that gametophyt es in temperate forests have a distinct ecology relative to sporophytes that gametophytes can be long-lived in natural and especially in vitro settings, and robust when dealing with abiotic stresses. Yet, the

PAGE 13

4 ecology of tropical gametophytes remains e ssentially unstudied. The goal of this dissertation is to examine multiple aspect s of fern ecology, focusing on the gametophyte generation.

PAGE 14

5 CHAPTER 2 GAMETOPHYTE ECOLOGY AND DEMOGRAPHY OF TROPICAL EPIPHYTIC AND TERRESTRIAL FERNS Introduction Ferns are conspicuous components of te mperate and especially tropical wet forests. Yet, general fern ecology is poor ly understood. Much early work was anecdotal or derived from studies and observations ma de from sporophytes or comments obscured in floristic inventories. A flurry of recent st udies attempted to describe both the patterns of fern sporophyte diversity and the causal re lations behind such patterns (Tuomisto and Ruokolainen, 1994b; Tuomisto and Dalberg, 19 96; Tuomisto, Poulsen, and Moran, 1998; Tuomisto and Poulsen, 2000; Jones et al., 2006 ; Watkins et al., 2006). These studies were critical in developing ecological models to better understand the biology of the fern sporophyte. Yet, focusing on sporophyte ecolog y, only told us a small part of fern ecology. Missing are studies on the ecol ogy of the free-living gametophyte. Ferns alternate between tw o independent generations: the haploid gametophyte and the diploid sporophyte. The gametophyte is a fundamentally different organism than the sporophyte. It lacks vascular tissue, produces rhizoids instead of true roots, has poorly developed to non-existent cuticles, and is co mparatively small. We know from early work that gametophytes can be more widespread a nd can grow in areas that are uninhabitable to sporophtyes. Yet, recruitment happens in the gametophyte generation, and the resulting sporophyte distributions de pend on gametophyte ecology.

PAGE 15

6 Gametophyte biology is complex, and ontogeny and morphology vary tremendously among species (Atkinson and Stokey, 1964; Nayar and Kaur, 1971). A common observation is that there are th e apparent fundamental differences in morphology and potentially longevity between ep iphytic and terrestria l species (Dassler and Farrar, 1997, 2001). Epiphytic species of ten produce gametophytes with diverse morphologies that are freque ntly capable of asexual reproduction and are potentially long-lived. Most terrestrial species are thought to produ ce the short-lived, textbook cordate thallus and exhibit little ability to reproduce asexually (but see Watkins and Farrar 2005). Little quantitativ e data have been generated to back up such longevity claims, and I have been unable to find a singl e paper that describes factors that influence the distribution and mortality of tropical gametophtyes. The goal of our study was to examine the cau sal mechanisms of the distribution of fern gametophytes and the demography of se veral tropical epiphytic, hemiepiphytic, and terrestrial species. We examined the distribut ion of epiphytic and te rrestrial gametophytes and ask what in situ factors control gametophyte establishment. Then we examined the gametophyte demography of 5 species from diffe rent habitat types, to assess gametophyte survival and recruitment rates and relate these to life history. Materials and Methods Study Site This study was conducted at La Selva Biologi cal Station (Heredia Province) in the Atlantic lowlands of northeastern in, Costa Ri ca. La Selva is a 1400 ha tropical wet forest having a mean annual rainfall of about 4,300 mm with peaks of precipitation in JuneJuly and November-December, and a drier period in March. Mean monthly rainfall

PAGE 16

7 nevertheless never falls below 150 mm in a ny month during the dry season based on long-term meteorological records. Gametophyte Transects In order to describe the occurrence of gametophytes in nature, 425 plots of 25 cm 25 cm were placed along 50 randomly chosen terrestrial 50m transects, and 425 25 cm 25 cm canopy plots were placed in 9 canopy trees. The total number of gametophytes (irrespective of identification) was counted and recorded. Each quadrat was coded for level of dist urbance: 0=undi sturbed (<5cm2 bare substrate); 1 = low disturbance (i.e. >5cm2 bare substrate); 2 = medium disturbance (>5cm2 bare substrate and substrate disturbed); 3= high disturbance (100% bare substrate and substrate turned over). Additionally, each quadrat with a dist urbance rating of Level 1 and above was coded for the type of damage when possibl e. To assess light environment, a digital hemispherical photograph was taken with a Ni kon Coolpix 950 digital camera (Melville, NY, U.S.A.) with a fisheye lens attachment, then analyzed using Gap Light Analyzer software (Frazer et al., 1999) to estimate the percentage of total light transmittance. Photos were taken 25 cm above each quadrat To determine the influence of light environment on density, both number of game tophytes and percent canopy transmittance were log transformed and analyzed by regression analysis. We used ANOVA to determine the influence of level of disturba nce on gametophyte density. Unless otherwise stated, all analyses were pe rformed with the computer program JMP version 5.01 (SASInstitute, 2005). Disturbance Plots To better understand the infl uence of disturbance and light on terrestrial fern establishment, 20 disturbance plots were established and monitored for gametophyte

PAGE 17

8 density at 5 months post estab lishment. All plots were established on the same soil type in primary forest, with 10 plots placed in low-light understory habitats and 10 placed in high-light canopy gaps of similar age. Each plot measured 1m2 and was divided into for 0.5 m 0.5 m subplots of increas ing disturbance that were similar in degree to those found in nature. The undisturbed treatment subplot acted as the control, and no leaf litter was removed. For low disturbance level, we re moved all leaf litter with no mechanical damage to the soil. The medium disturbance level was raked with a metal sand rake to disturb the first 5 cm of soil. The high dist urbance level was physica lly turned over with a shovel, to a depth of approximately 20 cm. Gametophyte density and diversity were recorded in the center 25 cm2 area. Litter-fall was removed from the disturbed plots weekly, and after 5 months, plots were assess ed for density (and when possible, diversity of gametophytes). Light environment was determined with di gital photography as discussed above. Determination of gametophyt e identity was difficu lt, and individuals were thus lumped into types that in act uality may represent multiple species. Types were assigned based on morphological character s that were identifiable by the use of a 10X or 20X hand lens and were identified a nd organized based on: trichome presence and type, rhizoid color, gametophyte shape, and the presence and morphology of gemmae. Identification of fern species from gamet ophytic characters was complicated and should be taken as a conservative estimate of act ual species richness. Only those gametophytes that were mature were counted. A 2X4 full factorial ANOVA was used to determine the effects of both light and disturbance inte nsity on gametophyte density and diversity. Demography In June of 2003, gametophytes from thr ee populations of each of 5 species (See table 2-2) were located and marked in the field. Marked individuals were checked once

PAGE 18

9 each month and followed for the next 15 months with one final census made at 25 months. No data were recoded during the 5-7th month. At each census, individuals were recorded as present, dead or missing, or as recruits into the sporophyte generation. When possible, individuals were code d for their cause of mortality. In the case of terrestrial species, ga metophytes were marked with a numbered aluminum nail; whereas, the epiphytic species were either marked with a nail or with a numbered tag attached to the substrate with copper wire. It was not possible in all cases to determine precisely the in itial age of marked gametophyt es. Therefore, individuals were chosen according to their initial size. Initial sizes were held constant within a species but differed among the species. Longevi ties were calculated as the time between the initial mark (treated as birth) and death of each gametophyte. Species were chosen to represent different functional types as discus sed below. A major flood event took place in month 11; individuals were sampled three da ys before the flood (for the regularly scheduled 11 month survey) and then three da ys after the flood to serve as an extra survey to determine the influence of fl ooding. The next sample period took place on the next corresponding survey day and was reco rded as the month 12 survey period. This allowed for more precise determination of mortality due to flooding rather than categorizing these individua ls in the unknown category. Gametophyte Survival Analysis As with many demographic studies, i ndividuals can leave the study by different avenues. Such absent samples were coded as right-censored data points (Hollander and Wolf, 1999). In this study, only those indivi duals that recruited into the sporophyte generation and those still alive at the end of the experiment (25 months) were recorded as censors. Censoring individuals reduces the samp le size of individuals at risk after the

PAGE 19

10 time of censorship. Censoring, therefore, redu ces the number of indi viduals contributing to the curve, and each death after a censored point represents a higher proportion of the remaining population. Subsequent deaths wi ll result in greater decreases in overall survivorship. Censorships th at occur early in the study have a greater effect on survivorship curves than those removed at later periods. Thus, the data from the survivorship curve after the first censor represent an estimate and not the actual survivorship of the population. In order to clarify the survival curves, we also plotted the cumulative proportion of individuals that recruited at each time interval. Gametophyte survival functions were es timated using non-parametric KaplanMeier product-limit survival func tions (Collett, 2003) These analyses were also used to estimate mean life span fo r each species. Log-rank 2 statistics were computed to test for homogeneity of the survival f unctions for all species. Weibu ll distributions were used to model survivorship functions a nd to calculate the parameters and The scale parameter is a measure of the degree of hazar d for the species; whereas, the shape parameter determines the degree of change in the hazard function over time. Large values of correspond to low hazard levels (i.e. gr eater survivorship) where low values equate to rapidly decaying su rvivorship. Large values of (i.e. >1) correspond to an increasing hazard rate that affects older in dividuals over younger individuals. With a <1 younger individuals are more likely to die within the period of the experiment. Results Transects A combined total of 2096 gametophytes were sampled with 329 recorded from canopy, 538 from low-trunk, and 1229 from terrestr ial habitats. Level of disturbance had a highly significant effect on the number of gametophytes in terrestrial habitats (Fig. 2-

PAGE 20

11 2b, r2=0.65, F=53.58, p<0.001) with greater percenta ges of gametophytes occurring in more disturbed habitats. Less than 1% of terrestrial gametophytes were found in areas without disturbance. The opposite trend was a pparent in canopy habita ts where level of disturbance had less influence on numb ers of gametophytes (Fig. 2-2a, r2=0.11, F=2.40, p=0.08) and greater percentages of gametophytes occurred in less dist urbed habitats (58% of canopy gametophytes were found in areas with no disturbance). In terrestrial transects, seven causes of disturbance were identified: leaf-litter removed, new and old root tip-ups from fallen trees, rotten logs erosion, branch falls, and animal causes (Fig. 2-2b). A total of six cau ses of disturbance were identified in canopy habitat: insect, branch falls epi-slides, physical damage, animal, and unknown causes (Fig. 2-2a). Identification of causes in the low-trunk habitat was difficult and thus was excluded from all analyses. In the case of te rrestrial species, recent root tip-ups harbored the greatest number of gametophytes (>50%). The disturbance categor y with the greatest percentage of gametophytes in canopy habita ts was animal disturbance with ~20%. In addition to disturbance, canopy openness (as a surrogate for light level) exhibited a positive effect on the number of terr estrial gametophytes (Fig. 2-3a, r2=0.423, p=<0.0001), but exhibited little influence on canopy gametophyte density (Fig. 2-3b, r2=0.001, p=<0.539). Disturbance Plots A total of 1247 gametophytes from 16 morpho-types were counted in the experimentally disturbed plot s. There were 6 non-unique mo rpho-types found in the low light treatment. There were a total of 16 types with 10 unique t ypes in the high light treatments. There was a significant and pos itive effect of both increasing light and disturbance and the interaction of light and disturbance on gametophyte density among

PAGE 21

12 the plots (Fig. 2-4, Table. 2-1). Likewise morpho-type richness was significantly influenced by both light and di sturbance but exhibited no sign ificant interaction (Fig. 2-4, Table.2-1). Demography All combined, 809 gametophytes from the fi ve species were marked and followed throughout the demography study. A total of 26 3 gametophytes were marked from the understory terrestrial species Danaea wendlandii The three populations of this species were all recorded in the unders tory of primary forests from sites that were at least 50m from trail sides. We marked 275 gametophytes of Pityogramma ebenea an abundant species often found in full to partial sun in dist urbed sites such as road and trail sides. All populations of this species were recorded from disturbed sites within the forests or in open areas away from trail sides. Sixt y-seven gametophytes of the understory hemiepiphyte Lomariopsis vestita were marked from small diameter trees in primary forests. Two canopy epiphytes were also marked: 98 from the high-light epiphyte Vittaria lineata and 106 from the mediumlight understory epiphyte Campyloneurum brevifolium Survival distribution functions varied significantly among species for the entire survey period (Table 2-2, log-rank 2 = 386.2, d.f. = 4, p < 0.0001). Campyloneurum brevifolium had the highest mean longevity a nd was 5 times that of the lowest: Pityrogramma ebenea (Fig. 2-5a, 2-6a). When combined, the epiphytic species: C. brevifolium, Lomariopsis vestita and Vittaria lineata had higher mean longevities than the terrestrial species (log-rank 2 = 212.3, d.f. = 1, P < 0.0001). In all cases >1, indicating an increasing hazard rate suggesting that older in dividuals are more likely to die than younger individuals over the study pe riod. Percent recruitment varied among the

PAGE 22

13 species, with C. brevifolium < V. lineata < D. wendlandii < L. vestita ~ P. ebenea The cumulative proportion of gametophytes recruiting varied for all species over the sampling time of the study (Fig. 2-6b). In itial recruitment was highest for both terrestrial species. More than 30% of gametophytes of P. ebenea had recruited between plot establishment and the first census. No additional individuals recruited beyond month eight. Initial recruitment was lower for D. wendlandii but increased throughout the study period. Recruitment was lowest for V. lineata and C. brevifolium with essentially no recruitment occurring after the third census up until the 25th month (Fig. 2-6b). The percent of gametophytes still alive at the end of the study also varied from 0% in P. ebenea to just over 70% in C. brevifolium (Fig. 2-5b) A total of 7 causes of mort ality, including an unknown cat egory, were surveyed in the field. Catastrophic habitat failure occurred when habi tats were over 95% of the individuals were destroyed. This happened when entire trees fell in the case of epiphytes or when hill sides collapsed with some te rrestrial species. Flooding also resulted in catastrophic failure, but was se parated as unique disturbance type because we were able to directly assess its influence (Fig 2-2a). Mi nor erosion also resulted in the loss of some individuals as did a massive flood in month eight of the study. F ungal attack, herbivory, and physical damage (as would occur from a br anch or rock fall th at physically removed individuals from the population) were also identifiable causes of mortality. The unknown category likely consisted of a contribution of all of these plus unidentifiable novel disturbances. Each cause resulted in diffe rent magnitudes of mortality, with some categories completely absent from some sp ecies and/or populations (Fig. 2-2). Most %

PAGE 23

14 mortality was attributable to unknown causes su ch as an individual simply missing from the population without any sign of disturbance, etc. Discussion Gametophyte Distributions The present study clearly demonstrates the importance of disturbance for gametophyte establishment. Even minor disturba nces that remove le af litter and turn up the soil can produce sites for gametophyte es tablishment. Disturbance that physically turns up soil not only produces an exposed and competition free habitat, is can also exposes the underlying spore bank and provide additional propagules that may further contribute to density and richness. Surpri singly, the maximum numb er of terrestrial gametophytes found in the lowest level of natu ral disturbance was three and the majority of the undisturbed sites had no established gametophytes. To my knowledge, disturbance has never been reported to be an important factor influencing gametophyte densit y or shaping species distri butions. However, studies on the gametophytes of temperate species have highlighted the importa nce of nutritional and edaphic safe sites for the gametophytes of Lorinseria (Woodwardia) areolata (Cousens, Lacey, and Scheller, 1988) and Blechnum spicant (Cousens, 1981). Little is known of other factors influencing ga metophyte distributions and a few temperate studies have produced mixed results suggesting that ga metophyte gender expression may influence gametophyte distribution (Klekowski, 1969; Cr ist and Farrar, 1983) while others have found no relationship (Holbrook-Walker a nd Lloyd., 1973; Greer and McCarthy, 1999). Apart from these studies, little is known of the influence of these characters on tropical fern gametophyte distributions. Numerous studies have however, examined factors behind sporophyte distributions and have fingered important roles of microclimate

PAGE 24

15 (Nobel, 1978) and water availability in dry sites (Marquez et al. 1997) and edaphic factors (Tuomisto and Ruokolainen, 1994a; Tuomisto and Poulsen, 1996; Tuomisto, Poulsen, and Moran, 1998) in we t tropical sites. None of these studies have directly addressed the role of disturbance. The nature of disturbed habitats creates a positive feedback for species that prefer disturbed sites. By their very nature such sites are unstable and result in increased mortality due to continued habitat erosion. Pityrogramma ebenea is perhaps the most common species of disturbed habitats at La Selva. The species produces large numbers of spores with high fecundity, and the gametophyt es can be found in virtually any habitat where disturbance is present, i.e. exposed road/trail cuts and the relatively dark understory (pers. obs.). The gametophytes of this species also germinate, grow, and recruit rapidly (Fig. 2-6). Ra pid development is necessary in species that occupy highly disturbed habitats. Catastrophic events are common in the hab itat of this species, and in two populations such disturbance resulted in th e near-complete habita t destruction and in continued habitat instability exacerbated by we t season rains. Epiphytic species are also subject to catastrophic distur bances; a single population of Vittaria lineata experienced this sort of disturbance following a large tree fall which resulted in 100% mortality of one of the study populations. However, these even ts seem relatively rare and epiphytic habitats tend to be more stable when compared to terrestrial habitats in this forest. Based on these data, terrestrial gametophytes simply do not establish in sites that are disturbance-free. However, there are varyi ng levels of tolerance and clearly different life histories in terrestrial species. Danaea wendlandii is a eusporangiate fern. The eusporangiates have many unique characters, but of particular interest for this study is the

PAGE 25

16 production of liverwort-like gametophytes that are several cell laye rs thick. Individual gametophytes are often large and more resistant to physical damage relative to the single layered leptosporangiate species (pers. obs.). Such tough game tophytes that occur in sites with minor disturbance confer longe vity. Indeed, the gametophytes of Danaea wendlandii exhibited 3 times the mean longevity and significantly less recruitment than those of P. ebenea Two populations of Danaea wendlandii fell in the flood zone, and while such disturbances are cl early part of the biology of th is species, this event resulted in lower mean longevities in this study. The eusporangiate biology of this species places it nearly opposite of P. ebenea in terms of life history strategy. There were also surprising differences be tween epiphytic and terrestrial species. One emergent difference between these two gr oups is the percent of gametophytes that survived but did not recruit. Ov er 70% of the gametophytes of C. brevifolium, and over 50% of both V. lineata and L. vestita were alive and un-recruited by the 26th month. This, compared to the less than 5% in Danaea wendlandii and 0% in Pityrogramma ebenea This observation highlights fundamentally diffe rent gametophytic life history strategies that have evolved in the two life forms. In fact, recent phylogenetic analysis of the ferns has revealed a recent split betw een terrestrial and epiphytic clades in the Eu-Polypodiales One of the defining gametophytic characters of the epiphytic clade is indeterminate and asexually reproducing gametophytes. Dassler an d Farrar (1997) have argued that such longevity and asexual reproduction is a mechan ism to encourage outcrossing in epiphytic species that may carry significant genetic lo ad. Long-lived thalli and thus genotypes can produce numerous archegonia over space and ti me to ensure fertilization of newly

PAGE 26

17 dispersed genotypes. Such differences in lif e history are significant and may have been critical in the radiation from terre strial species in to canopy habitats. Density and Species Richness Canopy gametophyte density is clearly more sensitive to disturbance, with nearly 58% occurring in undisturbed sites. Additiona lly, there was not a detectable relationship between gametophyte density and light environm ent as was shown for terrestrial density. In general, canopy light environm ents are significantly higher th an terrestrial sites and the lack of response to light in th e former is not surprising in a habitat where this light is not limiting. However, the canopy does experience temperature and humidity extremes and it is plausible that microclimate and water availa bility play larger ro les in these habitats relative to terrestrial sites. Such an observa tion would be in line with reports by Hietz and Briones (1998a) who demonstrated that w ithin canopy distribution of fern sporophytes was largely a function of species water relations. Quantification of gametophyte species/ morphotype richness in the natural transects was abandoned largely due to time constraints. However, there was a clear light-disturbance-density relationship for terrestrial species, and for this reason we examined these variables more completely in the terrestrial distur bance experiment. In one high disturbance plot, we counted six different morphotypes with 65% of the density dominated by Pityrogramma ebenea. There are numerous life hi story strategies in the ferns, and P. ebenea is a species with high germination a nd recruitment rates. Indeed, it is possible that all of the species that were encountered in the higher disturbance plots exhibited similar life histories. Clearly th ere are specific differences in gametophyte ecology that influence densities at a given si te. However, the near complete absence of gametophytes in undisturbed habitats suggests that disturbance may be critical to the

PAGE 27

18 majority, if not all terrestrial species regard less of life history. Such natural observations combined with experimental manipulati ons offer strong evidence that light and disturbance both act to st ructure terrestrial gamet ophyte density and richness. Unlike comparisons of seedling-adult distributions, fern gametophytes are completely and fundamentally different or ganisms from sporophytes. For this reason, gametophytes would not necessarily be expect ed to behave like sporophytes. Firstly, gametophytes are often, if not always more widespread than sporophytes (Peck, 1980; Peck, Peck, and Farrar, 1990). Such plasticity and simplicity in function may reduce the role that nutrients or othe r edaphic factors play in ga metophyte distribution. Secondly, gametophytes can be long lived with some indi viduals living decades in culture (Mottier, 1927; Walp, 1951) and in years field settings (pers. obs.). Additionally, gametophytes may be relatively sensitive to desicca tion and temperature changes; however, many studies have shown that gametophytes are re latively robust to envi ronmental stress (Sato and Sakai, 1980; Cousens, 1981; Sato and Sakai, 1981; Cousens, Lacey, and Scheller, 1988; Ong and Ng, 1998). While survival from stress and edaphic requirements may be important, for all species, emergence in litter free sites seems critical. Sporophyte Ecology Gametophyte and sporophyte distributions are related; however, the point of gametophyte establishment is not necessarily the point of mature sporophyte distribution. Epiphytic species are often associated with creeping rhizomatous growth. Such growth may allow a perfectly healthy gametophyte to produce sporophytes in less than optimal conditions. Such sporophytes may have the ab ility to then grow into more favorable habitats where they can reproduce. The resultant mature sporophyte distributions produced by such a strategy may obscure much of the species biology. Epiphytic species

PAGE 28

19 may also form gametophyte banks that are long lived and stress tolerant and like many seed banks, have the ability to wait for appropr iate conditions to appear before recruiting in to the sporophyte generation. This may be especially true for species whose gametophytes also have a means of vegeta tive reproduction. This ability is common in epiphytes (Atkinson and Stokey, 1964; Farrar, 1990; Farrar, 1998) and has been reported in some terrestrial species (Watkins a nd Farrar, 2005). Such ab ilities question the common assumption that gametophyte 'safe site s' limit the establishment of sporophytes in epiphytic species. This hypothesis may hold mo re merit with terrestrial species but is unlikely as important for epiphytic species. Conclusions Such apparent and fundamental differen ces in life history and the way that epiphytic and terrestrial life forms respond to disturbance and light provides evidence for adaptively meaningful variati on in life histories that ha s evolved in the two groups. Epiphytic species have evolve d in a high light, highly competitive, yet relatively stable matrix. Such environments reduce the light limitations encountered by terrestrial species, yet they incorporate closer contact with bryophytes. Dassler and Farrar (1997) have argued that differences in gametophyte morphology and as exual reproduction between epiphytic and terrestr ial species have largely evolve d due to pressures form bryophyte competition. Such changes may have only been possible in canopy habitats where disturbance is less intense. Radiation into canopy habitats required a suite of adaptive characters in both the gametophyte and spor ophyte generation. One major advantage of the canopy habitat is reduction in litter th at can cover developing gametophyte. While canopy habitats accumulate enormous amounts of organic matter (Cardelus and Chazdon, 2005), wind often removes a sign ificant proportion of leaves that land on internal and

PAGE 29

20 especially outer branches (Nadkarni and Ma telson, 1991). The presence of leaf litter may be the most important limiting factor to terrestrial species establishment as there are relatively few morphological or physiological pathways that would allow a species to survive under leaf litter. The mechanisms behind sporophyte distributions remain complicated as they clearly also rely on game tophyte ecology. This is further complicated by spore dispersal which may obscure the rela tionship between patter ns of distribution and habitat heterogeneity. Regardless of disper sal limitations or lack thereof, terrestrial sporophytes are largely elements of distur bance past. Additional work on gametophyte ecology will need to take into account ed aphic and microclimatic factors to better understand variables shaping the distribution of this magnificent group of plants.

PAGE 30

21 Table 2-1. Relationship of gametophyte density and richness with three levels of experimental disturbance and two light levels Gametophyte density Source DF F P > F Light 152.6430.000 Disturbance 318.5670.000 Light*disturbance 36.1520.001 Number of species Source DF F P > F Light 146.5220.000 Disturbance 36.6620.000 Light*disturbance 30.3640.779

PAGE 31

22Table 2-2. Demographic and survival analyses for the gametophytes of 5 fern species using the Wilcoxon test to compare survival distribution functions for different species Number of Gametophytes Survival Analysis Species Dead Censored Total Mean Months Std 2 P > X2 Campyloneurum brevifolium (Lodd. ex Link) Link 19 87 106 22.9 0.57 386.2 <0.0001 Danaea wendlandii Rchb. f. 168 95 263 12.3 0.56 Lomariopsis vestita E. Fmyn. 22 45 67 20.8 0.82 Pityrogramma ebenea (L.) Proctor 165 110 275 4.1 0.28 Vittaria lineata (L.) Sm. 40 58 98 16.6 1.08

PAGE 32

23 BLevel of Disturbance 0123 Number of Gametophytes 0 5 10 15 20 25 Level of Disturbance 0123 Number of Gametophytes 0 5 10 15 20 25 Ar2=0.11, F=2.40, p=0.08 r2=0.65, F=53.58, p<0.001a b b b a b b b Figure 2-1. Number of gametophytes counted and their relation to disturbance from natural transects in both A) Canopy habita ts and B) Terrestrial habitats. Here 0 represents no disturbance (< 5cm2 bare soil), 1 = low disturbance (i.e. > 5cm2 bare soil), 2 = medium disturbance (>5cm2 bare soil and soil disturbed), 3= high disturbance (100% bare soil and soil turned over).

PAGE 33

24 B Terrestrial HabitatsType of Disturbance NLOTRTERBFATTU Percent Gametophytes 0 10 20 30 40 50 60 A Canopy HabitatsType of Disturbance INBFESPDUNAT Percent Gametophytes 0 10 20 30 40 50 60 Figure 2-2. The percentage of fern gametophyt es as influenced by type of disturbance identified. A) In canopy habitats (IN: in sect damage, BF: branch fall, ES: epislide, PD: physical damage, UN: U nknown, AT: animal trail) and B) In terrestrial habitats (NL: no leaf litter, OT: ol d tip-up mound, RT: rotting tree/wood, ER: erosion, BF: branch fall, AT: animal trail, TU: recent tip-up mound)

PAGE 34

25 B Terrestrial HabitatsLog (Canopy Openess) 0.00.51.01.52.02.5 Log (Number of Gametophytes) 0.0 0.5 1.0 1.5 2.0 A Canopy HabitatsLog (Canopy Openess) 1.41.51.61.71.81.92.0 Log (Number of Gametophytes) 0.0 0.5 1.0 1.5 2.0 r2=0.000, p=0.085 r2=0.327, p=0.000 Figure 2-3. The relationship between canopy openness and gametophyte density for A. canopy and B. terrestrial species

PAGE 35

26 Level of Disturbance ControlLowMedHi Number of Gametophytes (m2) 0 200 400 600 800 1000 High Light Low Light Figure 2-4. Gametophyte densities as influenced by light and disturba nce in experimental plots. Control plots we re those that had <5cm2 of bare soil exposed. Low disturbance was created by only rem oving litter, medium plots were established by removing litter and raking soil to a depth of 5cm, high plots were created by removing litter and turni ng soil over with a spade to a depth of 20cm

PAGE 36

27 Mean Longevity (Months) 0 5 10 15 20 25 a b c d e Campyloneurum brevifolium Lomariopsis vestita Vittaria lineata Danaea wendlandii Pityrogramma ebenea Percent Gametophtyes Alive After 25 months 0 20 40 60 80 100 F=4.027, p=0.0337Campyloneurum brevifolium Lomariopsis vestita Vittaria lineata Danaea wendlandii Pityrogramma ebeneaa b b c d Figure 2-5. Mean longevity (months) A) and percent gametophytes still alive and unrecruited B) for the 25-m onth period of the study

PAGE 37

28 Time (Months) 051015242526 Proportion Surviving 0.0 0.2 0.4 0.6 0.8 1.0 1.2 051015242526 0.0 0.2 0.4 0.6 Time (Months)Proportion Recruiting Figure 2-6. Kaplan-Meier survivorship curves A) and proportion recruiting B) of 5 species of fern gametophytes over the 25-mont h study period. No data were collected during months 4-7 and 15-24 Campyloneurum brevifolium Lomariopsis vestita Vittaria lineata Danaea wendlandii Pityrogramma ebenea Pityrogramma ebenea Lomariopsis vestita Danaea wendlandii Vittaria lineata Campyloneurum brevifolium

PAGE 38

29 CHAPTER 3 COMPARATIVE DESICCATION TOLER ANCE OF TROPICAL FERN GAMETOPHYTES: ECOLOGICAL AND EVOLUTIONARY CONSEQUENCES Introduction Overwhelming evidence indicates that land plants evolved from simple aquatic algal ancestors (Bold, 1957; Niklas, 1997). The radiation of once-aqua tic plants onto dry land required the evolution of adaptive char acter suites that permitted life in what was and remains deadly dry air. To survive th is environment, plants have evolved two mechanisms of surviving desiccating condi tions. One mechanism is the avoidance of desiccation as demonstrated in most modern terrestrial plants and has been accomplished by the development of characters such as highly organized cuticles with effective stomatal control. Cacti of the dry deserts perhaps represent the pinnacle of avoidance with their succulent water storing stems, reduced leaves, and thick, well-developed cuticle. These characters and many other asso ciated with desiccation avoidance are all thought to be highly derived and early land pl ants most certainly did not have such structures. Early plants relie d on the unique mechanism of survival from desiccation, referred to as desiccation to lerance (DT). Desiccation tole rance has been differently defined; the most commonly accepted definition is survival of drying to equilibrium with surrounding air (Bewley, 1979). Such drying is more than sufficient to kill most any plant that relies on desiccation a voidance: at least 99% of the worlds vascular flora (Alpert and Oliver, 2002).

PAGE 39

30 Many of the characters that facilitated DT in ancestral land plants are still found in modern algae and bryophytes (Alpert, 2000 ; Oliver, Tuba, and Mishler, 2000; Alpert, 2005). Much as the cacti are to avoidance, the bryophytes are to tolerance. Fantastic stories exist in the literatur e demonstrating that some br yophytes can recover from 23 years of desiccation in herbaria (Alpert 2000 and references herein). Such remarkable abilities clearly have ecologi cal consequences and many studi es have shown that more desiccation tolerant bryophyt es are often associated with more xeric habitats. Consequences exist even within bryophyte species w ith one example being the overrepresentation of female gametophytes in dioecious bryophytes of xeric habitats. Such sex-based disparity is often related to greater DT in females relative to males (Stark 2005). Desiccation tolerance holds tremendous potential to influence the ecology of species (Deltoro et al., 1998; Csintalan, Pr octor, and Tuba, 1999; Robinson et al., 2000; Cleavitt, 2002). Apart from the linkage of DT to phyloge ny, there have also been demonstrated differences in desiccation tole rance of different generations, such as larval-adult in some invertebrates or gametophyte-sporophyte in some bryophytes. The ability and degree of desiccation tolerance in the different stages can be radically different. In some cases a high degree of tolerance may exist in one stag e but be absent in the other. For example, the gametophytes of some bryophytes exhibit a much greater degree of tolerance than sporophtyes (Proctor, 2000; Pr octor, 2001). Such variation also occurs invertebrates where the larvae of the fly Polypedilum vanderplanki exhibits greater desiccation tolerance relative to the adu lt stage (Watanabe et al., 2002). Within the vascular plants,

PAGE 40

31 desiccation tolerance of spor es and seeds is well known but much less is known about tolerance in lineages with two separate free-living stages. The only lineage of vascular plants to e xhibit two separate free-living generations is the pteridophytes. Of particular interest to desiccation tolerance are the morphological and physiological differences between these tw o stages in the ferns. The gametophyte is the point of gamete formation and fertilization and is small, lacks va scular tissue, and has a poorly developed to non-existent cuticle. The sporophyte produces spores and is thus the primary stage for dispersal; it has a well developed vascular system and a waxy cuticle complete with stomata. These diffe rences alone result in unique life-cyclemediated ecological strategies, especially as they relate to water relations. True desiccation tolerance in the sporophyte stage is known from and likely only exists in relatively few species (Gaff, 1987; Porembski and Barthlott, 2000). In a recent review on the subject, Proctor and Pence (2002) recorded that <1% (64 species) of the ferns studied exhibited DT and of those, 40 were Cheilantho id taxa that are commonly associated with desert-like habitats. Much less is known of species from tropical habitats, but DT has been recorded in genera as phylogenetically disparate as Asplenium and Polypodium (Kappen, 1964; Gaff, 1987; Proctor and Pence, 2002). The species in which it does occur are often extreme xerophytes living in de serts or other highl y exposed and dry environments. As with bryophytes, the appare nt degree of desiccation tolerance in the sporophyte stage has been linked with speci es ecological distribution (Harten and Eickmeier, 1987; Hietz and Bri ones, 1998b). Studies are still t oo sparse to determine the extent of this character in structuring populations.

PAGE 41

32 Studies on desiccation tole rance of the gametophyte ge neration of the ferns are fewer in number. Some of the earliest comm ents were made by Goebel (1900) regarding the ability of the buried tubercles of Annogramme chaerophylla to resume growth following dry spells. A similar observation was made by Cambell (1904) on the unburied gametophytes of Gymnogramme triangularis that appeared to have survived a dry California summer. The first evidence wa s generated by Picke tt (1913; 1914; 1931) who, through a series of desiccation experiments was the first to clearly show that the gametophytes of Asplenium rhizophyllum and A. platyneuron could recover growth following extreme desiccation. He also discove red that there was a greater degree of tolerance in A. Rhizophyllum a species of more exposed a nd drier habitats relative to A. platyneuron which is often confined to more me sic sites. This was the first link of desiccation tolerance in the gametophyte ge neration with sporophyt e distributions and species ecology. Picketts work has largely b een the last of its kind, and apart from anecdotal reports (Gilbert, 1970) and observati ons on the ability of the gametophytes of Pyrossia pilosellodes to recover from drought (Ong and Ng, 1998) nothing is known of the ability of fern gametophytes to tolerate desiccation and of thei r rates of recovery. The goal of this paper is to survey a br oad range of tropical fern gametophytes to determine the extent of desiccation tolerance in this phase of the fern life cycle. We first examine the ability of several species to reco ver from a single desic cation event. We then subject a select number of seve ral species of varying life hi stories to a more extensive repeated dry down cycles and drying intensit ies and relate the results to the ecological distributions of the species.

PAGE 42

33 Materials and Methods Spore Material and Growth Conditions Spore material from 12 species of varyi ng ecology was collected from La Selva Biological Station in the Atlantic lowlands of northeas tern Costa Rica at 37 masl. Fertile fronds were gathered in the field and put into glassi ne envelopes with tape-sealed seams. Envelopes with plant material inside were stored in an air-conditioned lab and allowed to dry under these conditions. Spores were brought back to the University of Florida where they were cultured. The grow th of a broad sampling of species with different ecologies required ex tensive experimentation with culture techniques, it was discovered that species grew best on a comb ination of organic soil collected from canopy trees at La Selva mixed with a small amount of vermiculite. Spores were sown on this medium into 60mm x 15mm Fisherbrand Petri pl ates. These plates were stored in sealed clear plastic containers (P ioneer Plastics Model 395-c, Dixon, KY). Cultures were exposed to 20mol m-2 sec-1 for 10hrs day-1 from GE fluorescent plant and aquarium 40watt growbulbs and watered with deionized water every 10-12 days. Desiccation Experiments For the initial survey experiment, 510 mature gametophytes (one gametophyte per Petri plate) of all 12 speci es (see Table 3-1) were allo wed to desiccate at a vapor pressure deficit (VPD) =1.3kPa (50% relati ve humidity) in a VPD controlled chamber that was constructed using one of the plasti c growth boxes connected via Bevline tubing to a Licor dew point generator (Model 610, Li ncoln, NE) set to a flow rate of 0.5 L min-1. Samples were allowed to dry for 45min in a constant vapor pressu re deficit and were removed from the box every 5min and placed in a Sartorious microbalance (Gttingen, Germany) where weight and a measurement of Fv/Fm was taken (see methods below). The

PAGE 43

34 samples were then placed back into the box. The volume of the box was relatively small, and a Hobo Pro RH/Temp Data Logger (Bourne MA) was used to verify that the container typically regained the 1.3kPa VPD within one min of the top being replaced. To evaluate recovery, upon completion of the desiccation treatment, samples were consecutively rehydrated by adding 5-10 drops of deionized water to the thallus. After 1 hr of rehydration, measurements of Fv/Fm were again made at 5min, 24hrs, and 48hrs post rehydration. Samples were then dried for 72hrs in a drying oven at 70C and weighed to determine gametophyte dry weight. Relative wa ter content was plotted against time and Fv/Fm. A second desiccation experiment was desi gned to test the effect of drying intensity on recovery of Fv/Fm. The gametophytes of Diplazium subsilvaticum and Phlebodium pseudoaureum, and Polypodium triserale were chosen to represent the extremes of tolerance from the initial desicca tion experiment and they were dried at three different intensities VPD=0.5kPa (20% RH ), VPD=1.3kPa (50% RH) and VPD=2.1kPa (80%RH) following the methods above. Gamet ophytes were kept at these VPD for 72hrs after which time they were rehydrated with deionized water and measurements of Fv/Fm were taken at 24, 48, and 72hr post rehydration. These values were related to the dark adapted value of Fv/Fm to determine the mean percent recovery. A third experiment was run to examine the influence of consecutive desiccation cycles on photochemical efficiency. Plant mate rial from six species was chosen from the survey experiment to represent the different recovery abilities. Thirty gametophytes were selected and ten were dehydrated for one, two, or three cycles at VPD=1.3kPa and kept at this level for 72 hrs using the methods desc ribed above. Material was then rehydrated

PAGE 44

35 with deionized water and measurements of Fv/Fm were again made at 24hrs, 48hrs, and 72hrs post rehydration. These values were re lated to the dark adapted value of Fv/Fm to determine the mean percent recovery. Chlorophyll-Fluorescence Measurements Variation in photochemical efficiency (Fv/Fm) was measured as the desiccation dependent change in ratio of variable and maximal fluorescence Fv/Fm where Fv is the difference between the maximum (Fm) and the minimum (Fo) fluorescence emissions (Mulkey and Pearcy, 1992; Horton, Ruban, a nd Walters, 1996) measured using an OptiSciences pulse modulated fluoromet er (Model OS-500, Hudson NH). Minimal fluorescence was measured under a weak pulse of modulating light over 0.8 s, and maximal fluorescence was induced by a sa turating pulse of light (8000 mol m-2 s-1) applied over 0.8 s. The parameter Fv/Fm was first measured after 20mins dark adaptation, and this measurement was taken as th e index of recovery. Dark-adapted FV/FM provides an estimate of the maximal quantum efficien cy of Photosystem II, which in unstressed material is generally around 0.76.83 (D emmig-Adams and Adams, 1992). Statistical Analysis For the initial desiccation survey, a series of regressions was run on arcsin square root transformed relative water content (RWC) data against time for each individual within a species to determine the rate of drying of gametophytes exposed to a VPD of 1.3kPa (50% RH) over the 45min time interval. The slopes of these regression lines were calculated to generate a speci es mean drying rate. These ra tes were then analyzed by a one way ANOVA followed by a post hoc Tukey s test to determine differences among species. Linear regression analysis was used to asses the influence both final relative and absolute water content at 45min and the slope of the individual drying curves on species

PAGE 45

36 recovery ability at 48h post rehydration. Percen t species recovery data were also arcsin square root transformed for these analyses The mean species dr ying rates, RWC, and absolute water content (AWC) at 45min were then plotted against each species mean percent recovery at 48h. Depre ssion in photochemical efficien cy was also graphed as a function of thallus RWC and AWC. To assess the influence of consecutive de siccation cycles (experiment 3) and VPD (experiment 2) on photochemical effici ency, a repeated measures ANOVA was performed with number of desiccation cycles or VPD and recovery time as the fixed main effects. Data were first examined for spheri city following the Mauchl y criterion. Pairwise comparisons were made across recovery times with Bonferroni-adjusted multiple t-tests. Klockars and Sax (Klockars and Sax, 1986 p. 38-39) recommend using the more stringent Bonferroni-adjusted multiple t-test when the number of planned comparisons is greater than the number of degrees of freed om for between-groups. In cases where data did not meet the sphericity criterion, p-va lues were adjusted using both GreenhouseGeisser and Huynh-Feldt methods based on the respective epsil ons (Scheiner and Gurevitch, 2001). Results Desiccation Survey For the initial desiccation survey, a series of regressions were run on each species to determine the change in relative water content (RWC) of gametophytes exposed to 1.32kPa (50% RH) over the 45min time interval All species exhibi ted rapid rates of thallus relative water content loss (Fig. 3-1A ) and absolute water content loss (data not shown). In all species regressi ons on the arcsin square root transformed data were linear (Table 3-1). In all cases, linea r regressions were used to calculate the slopes of species

PAGE 46

37 RWC and AWC drying curves to determine de siccation rates. The absolute size (mass based) of individual gametophyt es had little influence on the absolute rate of water loss (Fig 3-3a, r2=0.05, p=0.06). These rates varied si gnificantly among species with the fastest dry down in the terrestrial Thelypteris balbisii and slowest rates in the terrestrial Cyclopeltis semicordata and the epiphyte Polypodium triserale (Fig. 3-3A) Depression in photochemical efficiency (Fv/Fm) as gametophytes desiccated was non-linear with respect to RWC and varied among species (Fig. 3-1B). Species exhibited differential abilities to recover following desiccation (Fig 33B). This recovery ability was more clos ely related to the decay rate of RWC (r2=0.288, p<0.0001) when compared to the final RWC reached (r2=0.193, p=0.0008), the decay rate of AWC (r2=0.0001, p=0.81), or final AWC reached (r2=0.097, p=0.008) after 45min. Desiccation Rates The VPD of the different desiccation tr eatments significantly influenced the recovery abilities of both Diplazium subsilvaticum and Phlebodium pseudoaureum but had little influence on Polypodium triseriale. For the understory terrestrial D. subsilvaticum the ability to recover following the 2.12 and 1.3kPa VPD treatments was essentially non-existent. Additionally, the Fv/Fm values reached at these VPDs are suggestive of significant phot oinhibition and photodamage. The 0.53kPa treatment also depressed Fv/Fm but to a lesser degree and gametophytes exposed to this treatment exhibited clear recovery following rehydration. Phlebodium pseudoaureum exhibited relatively less depression in Fv/Fm a nd exhibited greater recovery than Diplazium subsilvaticum In all three VPD treatments, gamet ophytes exhibited recovery albeit with lower rates from the 2.12 and 1.32kPa treatments. Polypodium triserale exhibited

PAGE 47

38 remarkable Fv/Fm fidelity at all three ra tes with no significant degree of Fv/Fm depression at any desiccation in tensity (Fig. 3-5, Table 3-2). Desiccation Cycles Six species representing different life hist ories and desiccation tolerance from the initial survey were chosen and exposed to mu ltiple desiccation cycles (1, 2, or 3). In all cases excluding Polypodium triserale, percent recovery was gr eater following one versus two or three desiccation cycl es (Fig. 3-6 f, Table 3-3). Recovery ability was closely linked to species ecology with slow to no r ecovery following >1 desiccation cycle in the understory species that occur in more mesic habitats: Diplazium subsilvaticum, Adiantum latifolium, and Cyclopeltis semicordata Within this group, Cyclopeltis semicordata exhibited less inhibition and the greatest recovery followi ng a single desiccation cycle. With multiple cycles, all species were signifi cantly inhibited and there was little evidence of recovery following 2 and 3 cycles. Ther e was a much greater degree of recovery following 2 and 3 cycles in the species fr om more xeric habitats: the terrestrial Pityrogramma ebenea and the epiphytes: Phlebodium pseudoaureum, Polypodium triserale Unlike the other species in this xeric category, Polypodium triserale exhibited similar degrees of recovery from 1 and 2 cycl es but experienced a much slower recovery following the third de siccation cycle. Discussion In all species, fern gametophytes exhibi ted rapid rates of water loss (Fig 1a.). With poor control of transpiration and ineffi cient absorptive organs, gametophytes likely rely on water derived directly from the atmos phere and/or that which flows over them. As such, gametophytes face considerable variatio ns in water content throughout the day and must be able to withstand long periods of desicc ation. This is especially true of epiphytic

PAGE 48

39 gametophytes that can live for years (Watkins et al. Chapter 2). Th e only recourse that gametophytes have is to tolerate desiccation or perish. The initial survey produced a surprising degree of desiccation tolerance across many species in a gametophyte that is known to require water for fertilization and is thought to require humid conditions for su rvival. After exposure to a VPD of 1.3kPa (rH=50%) for 24 hours, all species exhibited greater than 50% recovery of the pretreatment Fv/Fm values and the majority had recovere d more than 70% of this value (Fig. 3-3). Gametophytes were dried to near constant state and thus match the definitions of desiccation tolerant by Bewley (1979) and Alpert and Oliver (2002). The species in this study are all tropical in origin and while they experience vari ous degrees of humidity in nature, some of the more exposed canopy trees for this same forest average VPD~3.0kPa at mid-day during the dry season (Carde lus and Chazdon, 2005). The gametophytes in this study experience VPD levels of 1.3kPa, but the value remains on the extreme end of what they typically experience. There was no significant difference in the absolute water loss rates based on gametophytes mass (Fig. 3-2) In natural settings, individuals will be exposed to daily variation in RWC and their ability to recover from low relative water content is crucial. The only apparent mech anism that gametophytes have to control water loss is an increase in size and perhaps alteration of thallus morphology (see below). Variation in VPD To further examine the influence of desiccation intensity on recovery of pretreatment photochemical efficiency, three species were exposed to three different desiccation intensities: roughly the everyday va por pressure deficient in nature (0.53kPa), that which is likely to reflect a typical drought event (1.32kPa) and an extreme value representing a VPD that species in this site rare ly if ever experience (2.12kPa). The

PAGE 49

40 results from this experiment demonstrated remarkable tolerance to desiccation intensity that is tightly linked to species ecology. Diplazium subsilvaticum is a low-light creek-side species and has little tolera nce of desiccation intensities below 0.53kPa (80% RH) (Fig 35A). The biggest decrease in Fv/Fm occurr ed in this species between 0.53 and 1.32kPa. The two epiphytes occur in similar habitats; and, Phebodium pseudoaureum is typical of open and exposed habitats such as roadsides and open clearings, Polypodium triseriale is often found in more highly exposed areas such as fence posts and tree trunks. The level of tolerance to desiccation in tensity was linked to the habitats with the most highly exposed Polypodium triseriale exhibiting essentially no sensitivity to desiccation intensity (Fig. 3-5C). These patterns enforce the notion that de siccation rates can influence the survival of certain species. In some bryophytes (Gaff, 1997; Alpert and Oliver, 2002; Alpert, 2005) and at least in the sporophtyes of the Hymenophy llaceae (Proctor, 2003), intermediate desiccation intensities that main tain intermediate RWCs are more likely to result in mortality compared to rapid dry dow ns. Alpert and Oliver (2002) have argued that one reason for this pattern is need fo r rapid yet organized metabolic shutdown that may not occur at slower desicc ation rates. Additional studie s that incorporate different drying intensities and longer time spent in these conditions are clearly needed. The link between DT and species distri butions corresponds cl osely with those reported for bryophyte species from xeric habitats exhibiting gr eater desiccation tolerance than those from mesic habita ts (Oliver, Mishler, and Qu isenberry, 1993; Deltoro et al., 1998; Proctor, 2001; Cleavitt, 2002; Alpert, 200 5). It also suggests that there may be some connection between gamet ophyte and sporophyte physiologies.

PAGE 50

41 Desiccation Cycles Not only do species experience different in tensities of desiccation in nature, they also experience multiple desiccation cycles throughout the day and/or growing season. The species in this study clearly exhibited di fferent abilities to cope with consecutive desiccation cycles at a VPD of 1.3kPa with species of more mesic habitat exhibiting little ability to cope with more than one cycle of desiccation (Fig. 3-6). The more mesic creekside Diplazium subsilvaticum was the most desiccation-se nsitive after one cycle and had the worst recovery whereas the more xeric Adiantum latifolium and Cyclopeltis semicordata had higher recoveries followi ng one cycle (Fig. 3-6 c&b). Polypodium triseriale, the species of more xeric habitats al so exhibited depre ssion in Fv/Fm, but recovery was largely independent of two desiccation cycles. One aspect that was common for all terrestrial species is the rela tive tolerance to a single desiccation cycle and the extreme depression caused by repeated cycles. Repeated desiccation cycles likely induce radical biological damage and recovery depends more on actual DNA repair and new protein synthesis than simple recovery of PSII function relate d to the release of excess excitation energy. Light and desiccation combined are often a deadly combination; and that gametophytes were returned to th e original culture c onditions upon rehydration could have resulted in increased photodamage. This was however similar to what species experience in nature and is likel y reflective of species biology. The ability of species to recover from de siccation was more closely related to the rate of drying rather than the final RWC (F ig 3-4). While gametophytes initially seem to have relatively few options to control the rate of thallus desiccation there was variation in rates among species exposed to identical dryi ng conditions (Fig 3-3a). For example, the terrestrial Thelypteris nicaraguensis a RWC decay rate that was more than twice as fast

PAGE 51

42 as the canopy epiphyte Polypodium triseriale The variation in de siccation rate was linked to gametophyte morphology (Fig. 3-4). Species with complex three-dimensional morphologies or those that te nded to produce prothallia ha irs exhibited significantly slower dry down rates than gametophytes th at are one dimensional and glabrous. This observation suggests a novel mechanism that gametophytes may employ to control water loss. There is limited internal capacitance in fern gametophytes, and in a manner similar to many bryophytes, fern gametophytes with even a minor degree of morphological complexity can hold external water. Such exohydric abilities may function in a natural setting to help slow the rate of water lo ss. The decrease in rate was more likely a combination of water vapor becoming trappe d in the folds and overlapping wings of more complex thalli and the increase of external boundary layer produced by gametophyte proliferations and hairs. Variation in gametophyte morphology especi ally in complexity has long been commented upon. Dassler and Farrar (1997, 2001) have speculated that complex morphologies found in many long-lived gametophyt es of epiphytic species have arisen from competition with bryophytes and as a mechanism to ensure outcrossing. One obvious trend across the taxa is that gametophyt es of species from drought-prone habitats such as those in epiphytic habitats, deserts, rock outcroppings, etc ., tend to produce thalli that often exhibit complex br anching, overlapping wings, pr oliferation, and hairs. Apart from this link between gametophyte morphology and ecology there is also a link between gametophyte morphology and phylogeny. Recent phylogenetic analysis of the ferns has revealed a split between terrestrial and epiphyt ic clades in the Eu-Polypodiales (Pryer, Smith, and Skog, 1995). The athyrioids, thel ypteroids, onocleoids, woodsioids and

PAGE 52

43 blechnoids are almost entirely terrestrial; pe rhaps less than 1% of the known species of this clade are true epiphytes. On the other hand, the dryopteroids, lomariopsoids, elaphoglossoids, oleandroids, davallioids, a nd polypodioids have ma ny epiphytic species; perhaps as many as 60-70% of the species in th is group as a whole are epiphytic (Fig 3-6) (R. Moran pers. comm.). Most epiphytic sp ecies exhibit complex morphologies whereas terrestrial species often less complex ones. Increases in th allus size and complex three dimensional morphologies may provide th e only water conservation mechanisms available to fern gametophytes. Complex water conserving morphology may have been critical in the radiation of ferns in to canopy and more exposed habitats. Conclusions The data presented in this paper show re markable desiccation tolerance in fern gametophytes. While all species exhibited r ecovery following an extreme desiccation event, the extent of recovery differed among species and was closely linked to the ecology of the species. The role of the fe rn gametophyte in controlling recruitment remains unclear, but these data suggest that gametophytes are relatively robust in dealing with desiccation especially in limited cycles. While desiccation intensity clearly influenced recovery, repeated cycles of de siccation were more likely to limit recovery and in the case of the most mesic species likely resulted in signif icant photodamage. The degree of recovery following desiccation and its relation to species ecology suggest that fern gametophytes exhibit adaptively meaningful variation in this ch aracter. It is likely that selective pressures ac ting on the gametophyte are la rgely responsible for the distribution of ferns and have played a major ro le in the evolution of ecological diversity within the group.

PAGE 53

44Table 3-1. Species and life form from the in itial desiccation survey, a se ries of regressions were r un on each species to deter mine the change in relative water content (slo pe) of gametophytes exposed to 1.32kPa ( 50% rH) over the 45min time interval. All species exhibited rapid and uncontrolled ra tes of thallus water loss. In all species regressions on the arcsin square root transformed data were linear (STE are st andard errors, RWC is relative water cont ent expressed as ((g fresh weightg dry weight)/g saturated weight g dry weight ))*100, AWC is absolute water content e xpressed as mg water / mg dry mass, % rec corresponds to percent recovery of initial pre-treatment dark adapted Fv/Fm values Species Life Form r2 p Slope STE RWC STE AWC STE %rec STE Adiantum latifolium Lam. Terrestrial 0.755 <0.0001 -1.891 0.101 7.375 2.138 1.340 0.164 0.579 0.116 Cyclopeltis semicordata (Sw.) J. Sm. Terrestrial 0.811 <0.0001 -1.359 0.074 34.255 3.836 5.660 0.820 0.897 0.019 Dennstaedtia bipinnata (Cav.) Maxon Terrestrial 0.939 <0.0001 -1.670 0.045 19.858 2.777 2.320 0.290 0.739 0.072 Diplazium subsilvaticum H. Christ Terrestrial 0.862 <0.0001 -1.512 0.084 31.880 5.680 3.640 0.980 0.705 0.041 Nephrolepis biserrata (Sw.) Schott Terrestrial 0.882 <0.0001 -1.419 0.094 34.858 4.298 5.540 1.170 0.823 0.056 Phlebodium pseudoaureum (Cav.) Lellinger Epiphyte 0.913 <0.0001 -2.033 0.059 9.646 1.449 2.530 0.500 0.678 0.055 Pityrogramma ebenea (L.) Proctor Terrestrial 0.878 <0.0001 -2.017 0.081 13.737 3.048 5.370 0.670 0.499 0.100 Polypodium triseriale Sw. Epiphyte 0.973 <0.0001 -1.279 0.203 49.021 8.097 8.640 1.471 0.807 0.065 Pteris altissima Poir. Terrestrial 0.834 <0.0001 -1.611 0.090 23.856 9.015 9.490 0.981 0.771 0.065 Thelypteris balbisii (Spreng.) Ching Terrestrial 0.816 <0.0001 -1.490 0.120 32.866 7.795 3.130 0.211 0.828 0.070 Thelypteris curta (H. Christ) C. F. Reed Terrestrial 0.711 <0.0001 -1.463 0.116 18.875 5.829 3.380 0.794 0.600 0.127 Thelypteris nicaraguensis (E. Fourn.) C. V. Morton Terrestrial 0.762 <0.0001 -2.199 0.061 19.254 3.512 3.690 0.370 0.555 0.087

PAGE 54

45Table 3-2. Fv/Fm recovery results from the repeated measures ANOVA for gametophytes exposed to three different desiccation intensities: 20%RH (~0.53kPa), 50%RH (~1.32kPa) and 80% RH (~2.12kPa). The gametophytes of Diplazium subsilvaticum are often found in the unde rstory, whereas those of Phlebodium ps eudoaureum, and Polypodium triserale were collected in the mid and exposed canopy respectively. Gametophytes were ke pt at the VPD levels for 48hrs after which time they were rehydrated with deionized water and measurements of Fv/Fm were taken at 24, 48, and 72hr post rehydration. These values were related to the dark adapted value of Fv/Fm to determine the mean percent recovery Diplazium striatastrum Lellinger df F p Mauchly 2 p VDP 2 38.41 <0.0001 0.739 3.24 0.662 ID(VPD) 12 1.86 0.0741 Recovery Time 2 7.32 0.0033 Recovery Time*VPD 4 6.39 0.0012 Phlebodium pseudoaureum (Cav.) Lellinger df F p Mauchly 2 p VDP 2 42.19 <0.0001 0.716 3.59 0.61 ID(VPD) 12 2.2 0.0482 Recovery Time 2 24.31 <0.0001 Recovery Time*VPD 4 7.81 0.0003 Polypodium triseriale Sw. df F p Mauchly 2 p VDP 2 1.46 0.2716 0.779 2.67 0.75 ID(VPD) 12 4.53 0.3531 Recovery Time 2 0.6 0.6352 Recovery Time*VPD 4 5.09 0.0041

PAGE 55

46Table 3-3. Fv/Fm recovery results from the repeated measures ANOVA for gametophytes exposed to 1, 2, or 3 desiccation cycles at 50% RH (VPD~ 1.32kPa). Gametophytes were ke pt at this level for 48 hrs. Material was then rehydrated with deionized water and measurements of Fv/Fm were again made at 24hrs, 48hrs, and 72hr s post rehydration. These values were related to the dark adapted value of Fv/Fm to determine the mean per cent recovery. Adjusted p-vaul es are G-G Greenhouse-Geisser and H-F Huynh-Feldt ad justed probabilities Adjusted p Diplazium striatastrum Lellinger df F p Mauchly 2 Adjusted p G-G H-F Rate 23.070.06490.6301 5.0810.07880.00430.0016 ID(Trt) 123.580.3931 Recovery Time 298.7<0.0001 GG = 0.73 Recovery Time*Trt 44.060.0118 HF = 0.944 Adiantum latifolium Lam. df F p Mauchly 2 p G-G H-F Rate 21.530.23780.691 4.060.13120.00730.0029 ID(Trt) 122.090.4982 Recovery Time 2324.53<0.0001 GG = 0.764 Recovery Time*Trt 45.440.0029 HF = 0.999 Cyclopeltis semi cordata (Sw.) J. Sm. df F p Mauchly 2 p G-G H-F Rate 2600.01<0.00010.706 3.8270.14750.0180.0094 ID(Trt) 121.980.5498 Recovery Time 27.330.0083 GG = 0.773 Recovery Time*Trt 42.430.753 HF = 1

PAGE 56

47Table 3-3. Continued Pityrogramma ebenea (L.) Proctor df F p Mauchly 2 p G-G H-F Rate 257.65<0.00010.857 1.6880.4298<0.0001<0.0001 ID(Trt) 120.560.792 Recovery Time 280.09<0.0001 GG = 0.875 Recovery Time*Trt 411.67<0.0001 HF = 1 Phlebodium pseudoaureum (Cav.) Lellinger df F p Mauchly 2 p G-G H-F Rate 298.81<0.00010.462 8.480.0143<0.0001<0.0001 ID(Trt) 120.380.8699 Recovery Time 2169.52<0.0001 GG = 0.65 Recovery Time*Trt 420.49<0.0001 HF = 0.818 Polypodium triseriale Sw. df F p Mauchly 2 p G-G H-F Rate 224.82<0.00010.492 7.780.020.00460.0019 ID(Trt) 121.650.548 Recovery Time 287.4<0.0001 GG = 0.664 Recovery Time*Trt 46.780.0008 HF = 0.839

PAGE 57

48 Relative Water Content 0 20 40 60 80 100 F v /F m 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Time (min) 01020304050 Relative Water Content 0 20 40 60 80 100 Thelypteris nicaraguensis Thelypteris curta Pityrogramma ebenea Adiantum latifolium Dennstaedtia bipinnata Pteris altissima Diplazium subsilvaticum Thelypteris balbisii Phlebodium pseudoaureum Nephrolepis biserrata Cyclopeltic semicordata Polypodium triseriale A B Figure 3-1. (a) Gametophyte dryi ng curves from 12 tropical fe rn species of different habitats. Species were exposed to a VPD to 1.32KPa (~50%RH) for 45 min. (b) Depression of photochemical efficien cy in the same gametophytes over a series of decreasing thallus water contents

PAGE 58

49 Dry Mass (mg) 0.00000.00050.00100.00150.00200.00250.00300.00350.0040 AWC Drying Slope (H20 mg / Dry Mass mg) -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 Figure 3-2. The rate of absolute water loss relative to gametophyte size as indexed by dry mass

PAGE 59

50 T h e R e f T h e N i c P i t E b e A d i L a t D e n B ip P t e A l t D i p S u b T h e B a l P h e P s e N e p B i s C y c S e m P o l P i n RWC Loss m-1 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 a a b b b b c c d e e fF=10.23, p<0.0001 Species T h e R e f T h e N i c P i t E b e A d iL a t D e n B i p P t e A lt D i p S u b T h e B a l P h e P s e N e p B i g C y c S e m P o l P i n Proportional Recovery 0.0 0.2 0.4 0.6 0.8 1.0 A B Figure 3-3. (a) Rate of thallus drying as cal culated from Figure 3-1 for 12 tropical fern species of different habitats. (b) Prop ortional recovery of the pre-treatment dark adapted value of Fv/Fm in these same species

PAGE 60

51 Change in AWC min-1 -0.5-0.4-0.3-0.2-0.10.0 Fraction Recovery of Pre-Treatment Fv/Fm 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Change in RWC min-1 -2.4-2.2-2.0-1.8-1.6-1.4-1.2-1.0 Fraction Recovery of Pre-Treatment Fv/Fm 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 TheRef TheNic PitEbe AdiLat PhePse PolPin CycSem NepBig TheAma DipSub PteAlt Orm PteAlt PitEbe PolPin TheNic PhePse CycSem AdiLat DipSub NepBig TheAma Orm TheRefA B Figure 3-4. Proportional recovery of the pr e-treatment dark adapted value of Fv/Fm and rate of thallus water loss expressed as (a) relative water content ((g fresh weightg dry weight)/g saturated we ight g dry weight))*100 and (b) absolute water content (g wet weight/g dry weight)

PAGE 61

52 Phlebodium pseudoaureum (Mid-Canopy) Dark24h48h72h Fv/Fm 0.0 0.2 0.4 0.6 0.8 2.12 Kpa (20%Rh) 1.32 Kpa (50% Rh) 0.53 Kpa (80% Rh) Diplazium subsilvaticum (understory)Time (h) Dark24h48h72h Fv/Fm 0.0 0.2 0.4 0.6 0.8 2.12 Kpa (20% Rh) 1.32 Kpa (50%r Rh) 0.53 Kpa (80% Rh) Polypodium triserale (Exposed Canopy) Dark24h48h72h Fv/Fm 0.0 0.2 0.4 0.6 0.8 Phlebodium pseudoaureum (Mid-Canopy)A B Ca b c a a b b b b a b c a b c a b b Figure 3-5. Fv/Fm recovery gra phs for gametophytes of (a) Diplazium subsilvaticum, (b) Phlebodium pseudoaureum (c) Polypodium triserale exposed to three different desiccation intensities : VPD~0.53kPa (20%RH), VPD~1.32kPa (50%RH), and VPD ~2.12kPa (80%RH). The gametophytes of Diplazium subsilvaticum are often found in the unde rstory, whereas those of Phlebodium pseudoaureum, and Polypodium triserale were collected in the mid and exposed canopy respectively. Gametophytes were kept at the VPD levels for 48hrs after which time they were re hydrated with deionized water and measurements of Fv/Fm were taken at 24, 48, and 72hr post rehydration. Pairwise comparisons were made acro ss recovery times with Bonferroniadjusted multiple t-tests

PAGE 62

53 244872 244872 244872 244872 244872 244872 Percent Recovery 0 20 40 60 80 100 244872 244872 244872 244872 244872 244872 0 20 40 60 80 100 244872 244872 244872 244872 244872 244872 0 20 40 60 80 100 244872 244872 244872 244872 244872 244872 Percent Recovery 0 20 40 60 80 100 Cycle 1 Cycle 2 Cycle 3 244872 244872 244872 244872 244872 244872 Percent Recovery 0 20 40 60 80 100 Recovery Time (hrs) Recovery Time (hrs) 244872 244872 244872 244872 244872 244872 0 20 40 60 80 100 a b b a b b a b b a b b a b b a b b a b c a b b a b c a b b a b c a b c a b c a ab b a a bA Diplazium subsilvaticum B Adiantum latifolium C Cyclopeltis semicordata D Pityrogramma ebenea E Phlebodium pseudoaureum F Polypodium triseriale Figure 3-6. Proportional Fv/Fm recovery results for gametophytes exposed to 1, 2, or 3 desiccation cycles at VPD~ 1.32kPa ( 50%RH). Gametophytes were kept at this level for 48 hrs. Material was th en rehydrated with deionized water and measurements of Fv/Fm were again made at 24hrs, 48hrs, and 72hrs post rehydration. These values were relate d to the dark adapted value of Fv/Fm to determine the mean percent recovery. Pa irwise comparisons were made within recovery times with Bonferroni-adjusted multiple t-tests. (a) Diplazium subsilvaticum (b) Adiantum latifolium (c) Cyclopeltis semicordata (d) Pityrogramma ebenea (e) Phlebodium pseudoaureum (f) Polypodium triseriale (see Table 3-1 for species habitat information)

PAGE 63

54 Figure 3-7. Morphology in fern gametophytes is diverse and is closely related to species ecology and phylogeny. Gametophytes of sp ecies from drought prone habitats such as those in epiphytic habitats, de serts, rock outcroppings etc. tend to produce thalli that ofte n exhibit complex branching, overlapping wings, proliferation and hairs; whereas, species from more buffered habitats have less ornamented and simple morphologies The athyrioids, thelypteroids, onocleoids, woodsioids and blechnoids are almost entirely terrestrial, perhaps less than 1% of the known species are true epiphytes. On the other hand, the dryopteroids, lomariopsoids, elaphoglo ssoids, oleandroid, davallioid and polypodioids have many epiphytic species perhaps as many as 60-70% of the species in this group as a whole are ep iphytic (R. Moran pers. comm.). Most epiphytic species exhibit complex mor phologies whereas terrestrial species often less complex ones. Such differe nces in morphology are significant and may have been critical in the radiatio n from terrestrial species into canopy habitats. A) Adiantum latifolium, B) Thelypteris sp.1, C) Thelypteris sp.2, D) Vittaria E) Campyloneurum

PAGE 64

55 CHAPTER 4 NITROGEN-15 NATURAL ABUNDANCE A ND NITROGEN USE STRATEGIES OF THE GAMETOPHYTES AND SPOROPHYTES OF TROPICAL EPIPHYTIC AND TERRESTRIAL FERNS Introduction A central goal of plant ecology is to develop a mechanistic understanding of species distributions in both space and time. On e important abiotic factor that is known to influence both plant performance and distribut ion is nitrogen. In Nlimited systems or in systems where N availability is heterogene ous, plants can compete for N in many ways, one of which is by partitioning this resour ce in both space and time or by uptake of different chemical forms of N. Such pa rtitioning may result in species coexistence through sorting along resource gradients which u ltimately structures species distributions. The partitioning and uptake of different ch emical forms of N has been receiving increased attention largely due to the revela tion that plants can circumvent the N cycle and directly uptake organic N, in the form of amino acids from the soil solution (Chapin, Moilanen, and Kielland, 1993a; Kielland, 1994, 1997; Lipson and Nasholm, 2001; Finzi and Berthrong, 2005). While it has been known for some time that plants can acquire organic N, early work in tundra and boreal ecosystems demonstr ated that a large component of an individuals N budget coul d be supported by direct uptake of organic relative to inorganic N (Chapin, Moilan en, and Kielland, 1993b; Kielland, 1994, 1997). Subsequent studies have shown that plants from a wide range of ecosystems can directly access organic N as an important part of th eir N nutrition (Lipson and Nasholm, 2001). Few studies, however, have examined the abilit y of plants from tropical wet forests to

PAGE 65

56 take up organic nitrogen and to our knowledge no studies have examined this ability in the ferns. The ferns pose a unique set of ecological c onstraints due to both the dispersal of tiny wind-blown spores and the occurance of an independent and free-living haploid gametophyte. The mineral nutrition of fern sporophytes is not well studied; however, evidence indicates that fern sporophytes behave as seed plant sporophytes when confronted with increased inorganic N (Prange and Ormr od, 1982; Walker and Aplet, 1994; Pillai and Ong, 1999). Less is known of the mineral nut rition of gametophytes and unlike sporophytes (which rely on well devel oped root systems), fern gametophytes are thought to rely primarily on rhizoid uptake of nutrients with possible uptake of water and nutrients across the thallus (Racusen, 2002). The ability of fern gametophytes to grow on different N forms was reviewed by Miller ( 1968), and there is limited evidence to show that the gametophytes of at le ast one species can grow well on specific mixtures of amino acids in the absence of inorganic forms. For plants other than ferns, changes in 15N values have been shown to occur through ontogeny and with increases in plant si ze (Zotz, 1997; Schmidt, Stuntz, and Zotz, 2001; Hietz and Wanek, 2003; Reic h et al., 2003; Zotz et al., 2004; Casper, Forseth, and Wait, 2005). In the case of hemiepiphytic plan ts, changes occur as a direct result of the connection of once aerial roots to terrestrial water and nut rient pools (Putz and Holbrook, 1989; Field, Lawton, and Dawson, 1996; Wanek et al., 2002). As ferns continue their life-cycle from gametophyte to sporophyte, radical changes in their ecophysiology, especially nutrient relations are likely to occur.

PAGE 66

57 The goal of this paper is to determine if species differ in their ability to take up different forms of nitrogen in both the game tophyte and sporophyte generations and to tie this to the natural abundance of 15N to determine if species access different nitrogen source pools. We then compare these data ac ross epiphytic and terr estrial species and a developmental series of a hemiepiphytic speci es to better understand the dynamics of nitrogen nutrition of different life forms. Material and Methods Study Site This study was conducted at La Selva Biol ogical Station of the Organization for Tropical Studies in Heredia Province, Costa Rica (10' N, 84' W). La Selva is a 1400 ha tropical wet-forest positioned in the Caribbean lowlands with an average monthly temperature of 25.8 C and annual rain fall of 4000 mm per year (Sanford et al., 1994). The site boasts a diversity of fern s with multiple species from epiphytic, hemiepiphytic and epiphytic life forms (Grayum and Churchill, 1987). Study Species The gametophytes and sporophytes of 10 species were field collected from 100X100m grids to control for differences in soil type in the terrestri al species and from the trees in the case of epiphytic species (Table 4-1). The following species were sampled: Adiantum latifolium Lam. is a terrestrial species that is common in disturbed areas in both primary and secondary forests. The sp ecies typically grows along trail sides and can be encountered under a wide ra ge of light and soil regimes. Danaea nodosa (L.) Sm. and Danaea wendlandii Rchb. F. are both eusporangiate ferns and as such have gametophytes that are several cell layers thick. Danaea

PAGE 67

58 wendlandii is perhaps the most common fern at La Selva and often grows on upland well drained sites and in dist urbed understory habitats. Danaea nodosa has similar habitat, but is more often associated with wetter site s becoming most abundant along creek sides. Diplazium subsilvaticum H. Christ is a terrestria l arborescent species that is common along creek banks and wetter areas in both primary and secondary forests. Lomariopsis japurensis (Mart.) J. Sm. and Lomariopsis vestita E. Fourn are both understory hemiepiphtyes. In both species, the gametophytes develop on the trunks of small trees and remain epiphytic throughout their life. Young sporophytes produce roots that grow down and contact the soil and re ly on the host tree for support. Adult plants never loose contact w ith the forest floor. Olfersia cervina (L.) Junze has been classifi ed as both a terrestrial and hemiepiphyte species. It is rest ricted to grow on soils with high organic content and is most commonly found growing on rotting logs, but can also grow on large tree trunks where sufficient detritus has accumulated. Elaphoglossum latifolium (Sw.) J. Sm. and Campyloneurum brevifolium (Lodd. Ex Link) Link are both canopy epiphytes with the former more often found in highly exposed portions of the canopy and often on bare bark. The latter species also occurs in exposed sites, but is most common in th e inner canopy rooted in canopy soil organic matter. Antrophyum lineatum (Sw.) Kaulf: An understory epiphyte that grows on the trunks of living trees. Both gametophytes and sporophytes are common in primary and secondary forests at La Selva.

PAGE 68

59 Isotopic Natural Abundance and 15N Labeled Uptake Foliar and gametophytic samples for natura l abundance of all nine species were field collected within a 3-wk period during June 2005. Collec tions were brought back to the lab where they were washed in deionized water to remove all soil and then dried at 60C for 48h. Samples were analyzed for nitr ogen concentration and isotope ratio using a Costech elemental analyzer coupled with a Finnigan Delta XL Plus continuous flow mass spectrometer at the University of Florid a, Gainesville. Based on repeat analyses of NIST peach leaves standard (SRM 1547; 15N 1.91), average 1s precision was 0.07 for 15N. In order to compare uptake of both or ganic and inorganic N forms we used excised root techniques (T reseder and Vitousek, 2001) in sporophtyes and whole plant uptake in gametophytes of Danaea wendlandii, Lomariopsis vestita, and Campyloneurum brevifolium Immediately prior to the uptake tria ls, plants were field collected and brought back to the lab where roots or gamet ophytes were rinsed with deionized water to remove soil and other debris. For sporophytes fine roots were selected, excised, and placed into a series of solutions co ntaining increasing concentrations of 15N labeled organic and inorganic N forms (see below) For gametophyte trials, 2-5 individual gametophytes of similar size and maturity were placed directly into a separate set of solutions. All samples were allowed to incubate for 60min in solutions containing 0 (deionized water only), 10, 50, 100, 300, and 500 mol concentrations containing only labeled NH4 +, NO3 -, (99 atom%), a cocktail of equa l proportions of the amino acids: Aspartic and Glutamic Acids, and Glycine (98 atom%), and a cock tail of all solutions (NH4 + + NO3 + the 3 amino acids). All solutions we re amended with 0.01 mol/L sucrose as an energy source and 0.5 mmol/L CaCl2 to maintain membrane integrity (Kielland

PAGE 69

60 1997). Immediately following the incubation tr ial, roots and gametophytes were removed and rinsed for 2min in a solution cont aining 1mmol/L KCl to remove any excess 15N from the external surfaces. The material was then dried at 60C for 73h, weighed, and ground for analysis of 15N using the same Finnigan Delta XL Plus continuous flow mass spectrometer. Nutrient Uptake Calculations 15N enrichment was calculated as F= [ T ( ASAB)]/ AF, where F is the weight of N derived from the 15N tracer, T is the total weight of N in the sample, AS is atom% excess 15N in the labeled sample, AB is atom% excess 15N in the natural abundance sample, and AF is atom% excess in the 15N tracer (Knowles and Blackburn, 1993). To calculate the kinetic uptake parameters of maximum uptake (Vmax) and the saturation constant (Km), we fitted the data to a Michaelis-Menten function. We also quantified the mols of N per g root dry mass against the solution concentration using V= Vmax S/ Km + S, where V is the velocity of uptake, and S is the concentration. The parameter Vmax is an estimate of the maximum uptake rate for a given ion a nd is controlled by the activity of membrane bound proteins specific to that ion. The va lue of experimental calculations of Vmax is that it gives an estimate of the roots/gametophyte s total capacity of ion uptake. The value Km is estimated to describe the affinity of specific membrane-bound proteins to a given ion and is related to the capaci ty to utilize low concentratio ns of this ion. Roots with lower Km values have higher affinities at low concentrations for the ion in question. Results 15N Natural Abundance and N concentration (mg g-1) Species differed significantly in 15N and N concentration (mg g-1) values (F=3.06, p=0.0053 and F=3.69, p=0.0009 respectively). There were also signifi cant differences

PAGE 70

61 within and between life forms ( 15N F=3.74, p=0.029; N concentration (mg g-1) F=10.66, p<0.0001), generations ( 15N F=15.52, p=0.0002; N concentration (mg g-1) F=57.65, p<0.0001), and a significant life fo rm by generation interaction ( 15N F=16.95, p<0.0001; N concentration (mg g-1) F=8.85, 0.0004) (Fig. 4-1, 4-2a). Within both hemiepiphytic and terrestrial species, the 15N values of gametophytes were more enriched F=64.93, p<0.0001 and F=3.84, p=0.0061) than that of sp orophtyes; whereas, the opposite was the case for the epiphytes (F=3.99. p=0.059). Across life forms, the gametophytes of terrestrial species were slightly depleted yet not significantly so when compared to epiphytic and hemiepiphytic species (F= 2.63, p=0.086). Greatest differentials were observed among the sporophtyes with hemiepiphyt es significantly more depleted than terrestrial species than epiphytic specie s (F=17.12, p<0.0001). Gametophytes exhibited significantly higher N concentration (mg g-1) relative to sporophytes within each life form and varied between life forms with epiphyt es and hemiepiphytes exhibiting higher N concentration (mg g-1) in gametophytes and sporophtyes than terrestrial species (Fig. 42b). To better understand ontogenetic shifts that occur in hemiepiphytic ferns, we measured variation in 15N natural abundance of different stages of Lomariopsis vestita. There was a clear series of increasing 15N enrichment from epiphytic gametophytes and sporophtyes to terrestrially-rooted adult sporophtyes (Fig. 4-3). 15N Labeled Uptake The gametophytes and sporophytes of both Danaea wendlandii and Campyloneurum brevifolium exhibited uptake capacity of both inorganic and organic forms of N. For both species uptake of NO3 in both gametophytes and sporophytes was limited. The gametophytes of both species had higher Vmax for all N forms than sporophytes. The gametophyt es and sporophytes of C. brevifolium had higher Vmax

PAGE 71

62 values for the amino acid mix followed by NH4 + and the all solution cocktail (Fig. 44a&b and Fig. 4-5a&b) and all solu tion cocktails. The gametophytes of D. wendlandii had highest Vmax values for the all solution cocktail where the Vmax for NH4 + and amino acid was similar (Fig. 4-5c). Km values exhibited considerab le variation both within and between species and generations. In all cases, Km was lowest and therefore affinity highest for NO3 relative to all other N forms (Fig. 4-6a-d). Discussion The results from the uptake studies indicate that ferns show preference for specific N forms and that they do so differently in sporophyte and gametophyte generations. In the case of the epiphytic C. brevifolium both gametophytes and sporophtyes exhibited high poten tial for uptake of amino acid N followed by inorganic NH4 + (Fig. 4-5a-b). Uptake potentials shifte d slightly in the gametophytes of the terrestrial D. wendlandii with uptake of amino acids and NH4 + essentially equal. The gametophytes and sporophytes of the epiphytic species exhibited higher uptake capacities for N-derived from amino acids relative to the terrestrial Danaea wendlandii Campyloneurum brevifolium is a midto low-canopy species that is frequently rooted in canopy soil; whereas, Danaea wendlandii is a species that is al ways rooted on mineral soil. The occurrence of species on such differe nt soil types is likely to produce radically different nitrogen nutrition and produce differe nt nutrient use strategies between such species. Canopy soil is fundamentally different from terrestrial soil in that the former is almost entirely organic. In a study on soil nutrien ts at La Selva, Cardelus and Mack (pers. com.) have shown that canopy soil organic ma tter had significantly greater bulk nitrogen, NH4 +, and dissolved organic nitrogen than terrestri al forest floor soils and that nitrogen

PAGE 72

63 mineralization rates were significantly lower in the canopy. They were also able to show that within canopy soils, the concentration of NO3 was low and NH4 + and dissolved organic nitrogen dominate this matrix. For this reason, it is not surprising that epiphytic species would exhibit pref erential uptake of NH4 + and organic nitrogen. Both the uptake rate (Vmax) and half saturation constants (Km) were higher for NH4 + than either amino acids or NO3 for the sporophytes of Danaea wendlandii (Fig 4-5, 4-6). In the gametophytes of this species, the values of Km for amino acids were several fold higher than those from the sporophtyes and from the gametophytes and sporophtyes of Campyloneurum brevifolium Uptake rates of amino acids and NH4 + from these same gametophytes were not signi ficantly different. These pa tterns indicate that NH4 + is an important N source for both, but especiall y, terrestrial sporophty es in this study. NO3 concentration within terrestrial soils at this site is high (Cardelus and Mack pers. com.). Nitrate is highly mobile and ferns must comp ete with microbes and other plants for this resource and may have partitioned uptake to NH4 and amino acids to avoid or lessen this competition. Such plants would not be highly invested in NO3 carriers and would be expected to exhibit higher a ffinities for this ion at lower concentrations; a result demonstrated by this study (Fig. 4-6). The importance of amino acids as compone nts of species N budgets is receiving increased attention and has been shown in species from tundra (Kielland, 1994), to temperate (Finzi and Berthrong, 2005) and s ubtropical (Schmidt and Stewart, 1999) ecosystems. We believe that our data are some of the first to demonstrate this ability in species from tropical lowland forests and clea rly the first to do so in the ferns. The significance of amino acids varies but clearl y makes up a critical component (>50% by

PAGE 73

64 some estimates) of species N budgets from tundra ecosystems (Kielland 1994). In many ways, canopy soil is functionally equivalent to tundra and boreal soils as it is organic in origin and potentially has high concentrations of free amino acids. As such, there may be convergence to greater investment in amino acid uptake in organic soils. The gametophytes of both species had higher uptake rates of all N forms excluding NO3 relative to sporophytes. There are fundamental differences in anatomy and morphology of these two stages of the lif e cycle and comparisons across stage must be made with care. Gametophytes produce primit ive yet functional rhizoids that likely aid in nutrient uptake (Smith, 1972a, 1972b), yet they may also take in nutrients via diffusion across cells of the thallu s (Racusen, 2002). This could result in much greater gametophyte surface area and transporter density compared to root surface area and result in greater uptake per unit ma ss in gametophytes. Expression of uptake rates on an area basis is made difficult as the gametophytes of both species develop complex three dimensional morphologies that make accura te determination of area difficult. There are a number of mechanisms that control the natural abundance 15N signatures in plant tissu es and for this reason, 15N signatures reflect a series of integrated fractionation events (Evans, 2001; Robinson, 200 1; Dawson et al., 2002). In spite of the complexities of interpreting natural abundance 15N signatures, such data provide evidence of a process when confounding variab les that influence fractionation can be identified or eliminated (Robinson, 2001). Differences in plant 15N signatures have been shown to be primarily related to 1) uptake of different N sources with distinct signatures (Robinson, 2001), 2) N availability and pl ant demand (Kolb and Evans, 2003), 3)

PAGE 74

65 mycorrhizal associations (Hobbie and Hobbie in press) 4) rooting depth (Nadelhoffer and Fry, 1994) or a combination of these events (Dawson et al., 2002). Both N availability and plant demand can have impacts on tissue 15N values through kinetic fractionations Kolb and Evans (2003) have shown that when N supply is greater that a plants assimilatory ability, plant tissues can become highly depleted in 15N relative to the source. This however, only occurs in ecosystems where external N concentrations are high and such kinetic fraction seems unlikel y to be responsible for the differences observed in our st udy. It is possible that mechan isms other than supply and demand drive kinetic fractiona tions differently in game tophytes and sporophtyes. Unfortunately, little is known of how root vs. rhizoid/th allus uptake differentially fractionate N. Ectomycorrhizal associations can result in large fractionation events (8-10 /oo) whereas fractionation caused by arbuscular my corrhizal symbioses se ems to have little effect on 15N values of host plants (Schmidt and Stewart, 2003). Both terrestrial and epiphytic leptosporangiate fern species have arbuscular mycorrhizal symbioses in the sporophtyes, but never in the gametophyt es (Gemma, Koske, and Flynn, 1992). The differences in natural abundance 15N signatures between gametophytes and sporophtyes are unlikely due to such associations. Plant uptake from source pools with different 15N signatures is a major contributor influencing tissu e values. Plants can take up both organic and inorganic nitrogen and each of these sources has different signatures: NO3 is highly depleted relative to NH4 + which can be more depleted that amino acids. In our data, gametophyte natural abundance 15N signatures are either significantl y depleted or equal to sporophyte

PAGE 75

66 values. The data from the uptak e experiment indicate that NO3 is not a major source of either gametophyte or sporophtyes N nutrition and that source alone is not responsible for the differences between gametophyte and spor ophtyes. Sources can however be derived from N species from enriched soil or highl y depleted atmospheric sources. Evidence indicates that epiphytes rely heavily on depleted atmosphe ric N sources (Hietz et al., 2002) and the result is major offset of 15N va lues when compared to terrestrially rooted plants (Watkins, unpublished data). This may also help explain the differences between gametophytes and sporophtyes. As throughfall leaches through the canopy to the forest floor, it may contain significan t concentrations of depleted N (Cardelus and Mack pers. com.) which may then be taken up more directly by gametophytes. Canopy epiphytic gametophytes also rely on depl eted N species and may exhib it greater direct atmospheric uptake than sporophytes. Our uptake data in dicate that overall uptake rate in gametophytes is much greater than that in sporophtyes. This may provide gametophytes greater opportunity for uptake of depleted poo ls that may be rapidly diluted by rainfall. Several studies have now shown that tissue 15N values from deep rooted individuals are more enriched relative to shallow rooted indivi duals (Nadelhoffer and Fry, 1988; Nadelhoffer and Fr y, 1994; Handley and Scrimgeour, 1997). Rooting depth is a potentially major contributor to the observed differences in 15N values between the surface rooted gametophytes and deeper sporophyte roots. One way to observe shifts in N sources is to follow ontogeneti c series and track changes in 15N values. The gametophytes of Lomariopsis vestita are epiphytes on understory trees and produce epiphytic sporopht yes with roots that grow down the trunk into the soil. The difference between game tophytes and young sporopht yes that were not

PAGE 76

67 attached to the soil were large with the latt er being more enriched (Fig. 4-3). This is possibly due to a combination of atmospheric uptake in gametophytes and quickly shifts to direct root uptake in spor ophtyes that may be more effici ent in tapping nutrients from more enriched soil pools. Terrestrially r ooted young and adult individuals would be predicted to exhibit enriched 15N signatures relative to ep iphytic individuals. Such plastic abilities are critical in the life of hemiepiphytes that face frequent water stress and a temporally heterogeneous nutrient environment. Conclusions The observed differences in foliar 15N values and evidence of differential uptake of N forms indicate that ferns can partition N by form. The preference of N form varied with greater preference on orga nic N in the epiphytic vs. terrestrial species and between gametophytes and sporophtyes. These data indi cate that there are different nutrient use strategies between the two life forms and between generations. I ndividuals that can circumvent mineralization in the N cycl e by direct amino acid uptake may have a tremendous competitive advantage relative to th ose relying on inorganic forms. What is critically needed are studies that incorpor ate dual labeled C and N isotopes to precisely determine if amino acids are hydrolyzed at the cell membrane or if they are taken directly into plants.

PAGE 77

68Species Life Form Distribution Adiantum latifolium Lam. Terrestrial Disturbed areas, high medium light Danaea nodosa (L.) Sm. Terrestrial Understory, low light Danaea wendlandii Rchb. F. Terrestrial Understory, di sturbed low light areas Diplazium subsilvaticum H. Christ Terrestrial Understory exposed mesic areas Lomariopsis japurensis (Mart.) J. Sm. Hemiepiphyte Understory secondary and primary forests Lomariopsis vestita E. Fourn Hemiepiphyte Understory secondary and primary forests Olfersia cervina (L.) Junze Hemiepiphyte-TerrestrialUnderstory, on mounds of organic matter or decaying trees Elaphoglossum latifolium (Sw.) J. Sm. Epiphyte Highly exposed, on bare bark in canopy Campyloneurum brevifolium (Lodd. Ex Link) Epiphyte Moderate light, rooted in soil, inner canopy Antrophyum lineatum (Sw.) Kaulf Epiphyte Understory secondary and primary forests Table 4-1. Species, life form and ecology for th e natural abundance a nd uptake experiments

PAGE 78

69 Danea nodosa -2 0 2 4 6 Adiantum latifolium 15N -2 0 2 4 6 Campylonerum brevifolium -2 0 2 4 6 Danea wendlandii 15N -2 0 2 4 6 Lomariopsis jauperensis -2 0 2 4 6 Lomariopsis vestita -2 0 2 4 6 Olfersia cervina 15N -2 0 2 4 6 Polytenium -2 0 2 4 6 Elaphoglossum latifolium -2 0 2 4 6 Sporophyte Gametophyte Sporophyte GametophyteF=2.131, p=0.218 F=51.40, p<0.0001 F=7.73, p=0.49 F=24.53, p=0.0006 F=6.803, p=0.0798 F=0.946, p=0.349 F=11.77, p=0.041 F=0.838, p=0.383 F=4.263, p=0.178 -4 -2 0 2 4 6 15N 15NF=33.01, p=0.0045 Microgramma reptans Figure 4-1. Sporophtyic and Gametophytic 15N natural abundance signatures of 10 tropical fern species. Tissue was field collected from 100X100m grids to control for differences in soil type in the terrestrial species and from the same trees in the case of epiphytic species

PAGE 79

70 EpiphyteHemi-EpiphyteTerrestrial 15N -1 0 1 2 3 4 5 Gametophyte Sporophyte EpiphyteHemi-EpiphyteTerrestrial mg g-1 0 1 2 3 4 5 Aa Aa Aa Ba Bb Bc Aa Aab Ab Ba Ba Bb Figure 4-2. Sporophtyic and Gametophytic (a) 15N natural abundance signatures and (b) N concentration (mg g-1) of epiphytic, terrestrial, and hemiepiphytic tropical fern species. Post hoc test were genera ted using Tukey tests. Capitol letters refer to within life-form whereas lower case letter refer to across life form comparisons

PAGE 80

71 d15N -1 0 1 2 3 4 5 Gametophytes Young Sporophytes Not Attached Young Sporophytes Attached Adult Sporophytes Attached F=20.34, p<0.0001 A AB BC C Figure 4-3. 15N natural abundance signatures from th e hemiepiphytic fern Lomariopsis vestita. Gametophytes of this species are completely epiphytic on understory trees. Young sporophtyes are initially produ ced that have true roots, but no connection to the forest floor at very early stages. Young sporophtyes eventually attach to the soil and root ing depth increases with adult plants

PAGE 81

72 DW GametophyteN Concentration (umol/L) 0100200300400500 Uptake Rate (umol g-1 h-1) 0 20 40 60 80 100 120 140 160 DW SporophyteN Concentration (umol/L) 0100200300400500 0 20 40 60 80 100 120 140 160 Amino Acid Mix All Solutions NH4 NO3 CB GametophyteN Concentration (umol/L) 0100200300400500 0 20 40 60 80 100 120 140 160 CB SporophyteN Concentration (umol/L) 0100200300400500 0 20 40 60 80 100 120 140 160 Uptake Rate (umol g-1 h-1)A B CD Figure 4-4. Uptake curves from 15N labeled solutions. Fine r oots were selected, excised, and placed into a series of solutions containing increasing concentrations of 15N labeled organic and i norganic N forms. For the gametophyte trials individual gametophytes of similar size a nd maturity were placed directly into a separate set of solutions. All samples were allowed to incubate for 60min in solutions containing only 15N labeled NH4 +, NO3 -, (99 atom%), a cocktail of equal proportions of the amino acids: Aspartic and Glutamic Acids, and Glycine (98 atom%), and a cocktail of all solutions (NH4 + + NO3 + the 3 amino acids)

PAGE 82

73 Campyloneurum brevifolium Gametophyte Nitrogen Form NH4NO3AAAll Vmax (mol N g dw-1 h-1) 0 50 100 150 200 Nitrogen Form NH4NO3AAAll Vmax (mol N g dw-1 h-1) 0 50 100 150 200 ANitrogen Form NH4NO3AAAll Vmax (mol N g dw-1 h-1) 0 50 100 150 200 ANitrogen Form NH4NO3AAAll Vmax (mol N g dw-1 h-1) 0 50 100 150 200 A B C AF=83.34, p<0.0001 Campyloneurum brevifolium Sporophyte F=74.92, p<0.0001 Danea wendlandii Gametophyte Danea wendlandii SporophyteA B C DF=13.0, p=0.0019 F=61.55, p<0.0001B A C B B A A A A B C D Figure 4-5. Uptake saturation values (Vmax) of each N form derived from MichaelisMenten functions of the data from Fi g. 4-2, error bars ar e standard errors. Campyloneurum brevifolium is a mid-canopy epiphyte of exposed habitats; Danaea wendlandii is a low-light understory terrestrial species

PAGE 83

74 Nitrogen Form NH4NO3AAAll Km (mol L-1) 0 50 100 150 200 250 300 Nitrogen Form NH4NO3AAAll Km (mol L-1) 0 50 100 150 200 250 300 Nitrogen Form NH4NO3AAAll Km (mol L-1) 0 50 100 150 200 250 300 Nitrogen Form NH4NO3AAAll Km (mol L-1) 0 50 100 150 200 250 300 Campyloneurum brevifolium Gametophyte Campyloneurum brevifolium Sporophyte Danea wendlandii Gametophyte Danea wendlandii SporophyteA A B AF=75.59, p<0.0001A A B AF=27.12, p=0.0002A C B AF=63.99, p<0.0001A B B BF=78.57, p<0.0001A B C D Figure 4-6. Uptake saturation values (Km) of each N fo rm derived from MichaelisMenten functions of the data from Fig. 4-2, error bars are standard errors. Campyloneurum brevifolium is a mid-canopy epiphyte of exposed habitats; Danaea wendlandii is a low-light understory terrestrial species.

PAGE 84

75 CHAPTER 5 CONCLUSIONS This dissertation is one of the first attempts to examine and apply modern ecological and ecophysiological techniques to the study of fern gametophyte ecology. This work has demonstrated that fern gametophytes can be extremely long-lived in situ and that there are differences in factors that influence the distribu tion and demography of epiphytic and terrestrial ferns. Differences in life history a nd the way that epiphytic and terrestrial life-forms respond to disturban ce and light provide evidence for adaptively meaningful variation in life histories that has evolved in the two groups. Epiphytic species have evolved in a high light, highly competitive, yet re latively stable matrix. Such environments reduce the light limitations en countered by terrestrial species, yet they incorporate closer contact w ith bryophytes. These habitat mediated conditions may be largely responsible for the obser ved variation in longevity. Dassler and Farrar (1997) have argued that differences in gametophyte longevity between epiphytic and terrestrial species have largely evolved due to pressures from the genetic consequences of intergametophytic selfing. Asexually repr oducing indeterminate gametophytes of many epiphytes can produce large and long-lived clones. Such clones greatly increase the longevity of individua l genotypes which is hypothesized to increase the chance of outcrossing. The data gene rated from chapter one clearly show that epiphytic gametophytes are significantly long er lived than terrestrial species. One remarkable discovery is that epiphytic gametophytes can live for years where even

PAGE 85

76 terrestrial species on relativel y stable substrates rarely lived beyond 6 months. I have observed gametophytes of some understory epiphytes that are over 6 years old. Are the differences in longevity between epiphytes and terrestrial species related to some adaptively meaningful variation betw een the life-forms as has been hypothesized, or is it simply a result of some intrinsic in stability of epiphytic vs terrestrial habitats? The answer to this question remains elusiv e as there is no clea r understanding of the differences in disturbance between epiphytic an d terrestrial habitats. The data in my study would indicate that epiphytic habitats are far more stable than terrestrial habitats. This clearly needs to be establis hed and a much greater survey of demography needs to be undertaken to better understand the extent of the differences that I have reported. Gametophytes that can live for months and espe cially those that li ve for years have to cope with stress associated with extr eme abiotic variation. The second part of my dissertation has revealed a surprising and completely unexpected degree of extreme gametophyte stress tolerance to desiccation acr oss several species. All species surveyed exhibited more desiccation tolerance than cu rrent pteridological dogma would suggest. In addition, such tolerance was clearly linked to species sporophyte eco logy with those from drought prone habitats, such as the epiphytes in the study, exhibiti ng greater degrees of tolerance compared to those in mesic habita ts. Epiphytic species were also robust in dealing single dry down events and exhibited significantly greater recovery following extreme desiccation intensities and multiple de siccation cycles compared to more mesic terrestrial species. Not only ar e epiphytic species longer-lived, they are also considerably more desiccation tolerant: two characters that are likely connected.

PAGE 86

77 Stress tolerance has been show to stru cture some bryophyte communities (Cleavitt, 2002) but I have been unable to find any additi onal work to suggest that ferns are sorting along clines of stress tolera nce. Much additional work needs to be completed on gametophyte physiology to truly understand th e role that gametophytic longevity and desiccation tolerance plays in sorting species. Additional studies need to include combinations of temperature and light st ress with desiccation to develop a better understanding of the interaction of these char acters and the relative tolerance of more species. This work also has basic science appl ications beyond ferns and can be applied to aspects directly related to ge netic engineering of desiccati on tolerance in crop plants. One factor that seems closely tied to so rting of terrestrial fern sporophtyes are edaphic factors (Tuomisto, 1998, Tuomisto, 1994, Tuomisto, 2002). The role that nutrients play in shaping fern gametophyte distributions is virtually unknown. My work on nitrogen relations of tropica l fern gametophytes has reveal ed unexpected versatility in nitrogen acquisition between both gametophytes and sporophtyes and between epiphytic and terrestrial species. Of great significan ce was the discovery that ferns can partition nitrogen by form and have the ability to take up amino acids and use them as an important component of their nitrogen budgets. The importan ce of amino acid uptake as critical components of species N budgets is currently receiving in creased attention and has been shown in species from tundra (Kielland, 1994), to temperate (Finzi and Berthrong, 2005) and subtropical (Schmidt and Stewart, 1999) ecosystems. Nitrogen associated with amino acids is clearly impor tant in both the N cycle of tropical fern gametophytes and sporophtyes and such flexib ility in accessing different nitrogen forms may provide species with differential comp etitive abilities result in one mechanism by

PAGE 87

78 which species are sorted along nut rient gradients. What is cr itically needed are studies that incorporate dual labeled C and N isotopes to precisely determine if amino acids are hydrolyzed at the cell membrane or if they are taken directly into plan.

PAGE 88

79 LIST OF REFERENCES ALPERT, P. 2000. The discovery, scope, and puzzle of desiccation tole rance in plants. Plant Ecology 151: 5-17. ______. 2005. The limits and frontiers of desiccation-tolerant life. Integrative and Comparative Biology 45: 685. ALPERT, P., and M. J. OLIVER. 2002. Drying without dying. In M. Black and H. W. Pritchard [eds.], Desiccation and Surviv al in Plants, 3-43. CABI Publishing, Oxon. ATKINSON, L. R.,and A. G. STOKEY. 1964. Comparative morphology of the gametophytes of homosporous ferns. Phytomorphology 14: 51-70. BEWLEY, J. D. 1979. Physiological-aspects of desiccation tolerance. Annual Review of Plant Physiology and Plant Molecular Biology 30: 195-238. BOLD, H. C. 1957. Morphology of Plants. Harper and Row, New York. BOWER, F. O. 1923. The Ferns. Cambridge University Press, Cambridge. CARDELUS, C. L., and R. CHAZDON. 2005. Inner-crown microenvironments of two emergent tree species in a lowland wet forest. Biotropica 37: 238-244. CASPER, B. B., I. N. FORSETH, and D. A. WAIT. 2005. Variation in carbon isotope discrimination in relation to plant pe rformance in a natural population of Cryptantha flava. Oecologia 145: 541-548. CHAPIN, F. S., L. MOILANEN, and K. KIELLAND. 1993a. Preferential use Of organic nitrogen for growth by a non mycorrhizal arctic sedge. Nature 361: 150-153. ______. 1993b. Preferential Use of Organic Nitr ogen for Growth by a Non-Mycorrhizal Arctic Sedge. Nature 361: 150-153. CHIOU, W.-L., and D. R. FARRAR. 1997. Comparative gametophyte morphology of selected species of the family Polypodiaceae. American Fern Journal 87: 77-86.

PAGE 89

80 CLEAVITT, N. L. 2002. Stress tolerance of rare and common moss species in relation to their occupied environments a nd asexual disper sal potential. Journal of Ecology 90: 785-795. COUSENS, M. I. 1979. Gametophyte ontogeny, sex expr ession, and genetic load as measures of populat ion divergence in Blechnum spicant. American Journal of Botany 66. ______. 1981. Blechnum spicant : habitat and vigor of optim al, marginal, and disjunct populations and field observ ations of gametophytes. Botanical Gazette 142: 251258. ______. 1988. Reproductive strategies of pteridophytes. In J. L. Doust and L. L. Doust [eds.], Plant Reproductive Ecology: Patte rns and Strategies, 307-328. Oxford University Press, New Youk. COUSENS, M. I., D. G. LACEY, and E. M. KELLY. 1985. Life-History Studies Of Ferns A Consideration Of Perspective. Proceedings Of The Royal Society Of Edinburgh Section B-Biological Sciences 86: 371-380. COUSENS, M. I., D. G. LACEY, and J. M. SCHELLER. 1988. Safe sites and the ecological life history of Lorinseria areolata American Journal of Botany 76: 797-807. CRIST, K. C., and D. R. FARRAR. 1983. Genetic load and lo ng-distance dispersal in Asplenium platyneuron. Can. J. Bot. 61: 1809-1814. CSINTALAN, Z., M. C. F. PROCTOR, and Z. TUBA. 1999. Chlorophyll fluorescence during drying and rehydration in the mosses Rhy tidiadelphus loreus (Hedw.) Warnst., Anomodon viticulosus (Hedw.) Hook. & Ta yl. and Grimmia pulvinata (Hedw.) Sm. Annals of Botany 84: 235-244. DASSLER, C. L., and D. R. FARRAR. 1997. Significance of form in fern gametophytes: clonal, gemmiferous gametophytes of Callistopteris baueriana (Hymenophyllaceae). International journal of plant sciences 158: 622-639. ______. 2001. Significance of gametophyte form in long-distance colonization by tropical, epiphytic ferns. Brittonia 53: 352-369. DAWSON, T. E., S. MAMBELLI, A. H. PLAMBOECK, P. H. TEMPLER, and K. P. TU. 2002. Stable isotopes in plant ecology. Annual Review of Ecology and Systematics 33: 507-559. DELTORO, V. I., A. CALATAYUND, G. GIMENO, and E. BARRENO. 1998. Water relations, chlorophyll fluorescence, and membrane permeability during desiccation in bryophytes from xeric, mesic and hydric environments. Canadian Journal of Botany 76: 1923-1929.

PAGE 90

81 DEMMIG-ADAMS, B., and W. W. ADAMS. 1992. Photoprotection and Other Responses of Plants to High Light Stress. Annual Review of Plant Physiology and Plant Molecular Biology 43: 599-626. DEMMIG-ADAMS, B., and W. W. ADAMS III. 1992. Photoprotection and other responses of plants to high light stress. Annu. Rev. Plant Physiol. Plant Mol. Biol. 43: 599-626. EVANS, R. D. 2001. Physiological mechanisms influencing plant nitrogen isotope composition. Trends in Plant Science 6: 121-126. FARRAR, D. R. 1967. Gametophytes of four tr opical fern genera reproducing independently of their sporophytes in the southern Appalachians. Science 155: 1266-1267. ______. 1971. The biology of ferns with asexua lly reproducing gametophytes in the eastern United States. Ph.D. Dissertati on. Ph.D. Dissertation, University of Michigan, Ann Arbor, MI. ______. 1990. Species and evoluiton in ase xually reproducing independent fern gametophytes. Systematic Botany 15: 98-111. ______. 1998. The tropical flora of rockhouse cli ff formations in the eastern United States. Journal of the Torrey Botanical Club 125: 91-108. FIELD, T. S., R. O. LAWTON, and T. E. DAWSON. 1996. Comparative nutrient relations in canopy-rooted and ground-rooted Didymopa nax pittieri Marchal (Araliaceae) hemiepiphytes in a wind-exposed tropical montane rain forest. Biotropica 28: 774-776. FINZI, A. C., and S. T. BERTHRONG. 2005. The uptake of amino acids by microbes and trees in three cold-temperate forests. Ecology 86: 3345-3353. FRAZER, G. W., C. D. CANHAM, P. SALLAWAY, and D. MARINAKIS. 1999. Gap Light Analyzer. Simon Fraser University & In stitute for Ecosystem Studies, Burnaby, British Columbia, Canada & Millbrook, New York. GAFF, D. 1997. Mechanisms of desiccation tolerance in resurrection plants. In A. S. Basra and R. K. Basra [eds.], Mechanisms of environmental stress resistance in plants, 43-58. Harwood Academic Publishers, London. GAFF, D. F. 1987. Desiccation tolerant plants in South America. Oecologia (Berlin) 74: 133-136. GEMMA, J. N., R. E. KOSKE, and T. FLYNN. 1992. Mycorrhizae in Hawaiian pteridophytes: occurrence and evolutionary significance. American Journal of Botany 79: 843-852.

PAGE 91

82 GILBERT, O. L. 1970. Biological flora of the British Isles. Dryopteris villarii (Bellardi) Woynar ( Aspidium rigidum .; Lastrea rigidia (Sw.) C. Presl). GRAYUM, M. H., and H. W. CHURCHILL. 1987. An introduction to the Pteridophyte flora of Finca La Selva, Costa Rica. American Fern Journal 77: 73-89. GREER, G. K., and B. C. MCCARTHY. 1999. Gametophytic plasticity among four species of ferns with contrasti ng ecological distributions. International Journal of Plant Sciences 160: 879-886. HANDLEY, L. L., and C. M. SCRIMGEOUR. 1997. Terrestrial plan t ecology and N-15 natural abundance: The present limits to interpretation for uncultivated systems with original data from a Scottish old field, Advances in Ecological Research, Vol 27, 133-212. HARTEN, J. B., and W. G. EICKMEIER. 1987. Comparative Desiccation Tolerance of 3 Desert Pteridophytes Response to Long-Term Desiccation. American Midland Naturalist 118: 337-347. HIETZ, P., and O. BRIONES. 1998a. Correlation between wa ter relations and withincanopy distribution of epipytic fe rns in a Mexican cloud forest. Oecologia 114: 305-316. ______. 1998b. Correlation between wa ter relations and within -canopy distribution of epiphytic ferns in a Mexican cloud forest. Oecologia 114: 305-316. HIETZ, P., and W. WANEK. 2003. Size-dependent variation of carbon and nitrogen isotope abundances in epiphytic bromeliads. Plant Biology 5: 137-142. HIETZ, P., W. WANEK, R. WANIA, and N. M. NADKARNI. 2002. Nitrogen-15 natural abundance in a montane cloud forest canopy as an indicator of nitrogen cycling and epiphyte nutrition. Oecologia 131: 350-355. HOLBROOK-WALKER, S. G., and R. M. LLOYD. 1973. Reproductive biology and gametophyte morphology of the Hawaiian fern genus Sadleria (Blechnaceae) relative to habitat di versity and propens ity for colonization. Botanical Journal of the Linnean Society 67: 157-174. HOLLANDER, M., and D. A. WOLF. 1999. Nonparametric statistical methods. John Wiley and Sons Inc., New York. HOLTTUM, R. E. 1938. The ecology of tropical pteridophytes. In F. Verdoorn [ed.], Manual of pteridology., 433. M. Nijhoff, The Hague.

PAGE 92

83 HORTON, P., A. V. RUBAN, and R. G. WALTERS. 1996. Regulation of light harvesting in green plants. Annual Review Of Plant Physiol ogy And Plant Molecular Biology 47: 655-684. JONES, M. M., H. TUOMISTO, D. B. CLARK, and P. OLIVAS. 2006. Effects of mesoscale environmental heterogeneity and dispersal limitation on floristic variation in rain forest ferns. Journal of Ecology 94: 181-195. KAPPEN, L. 1964. Untersuchungen uber die Jahr eslauf der Frost-, Hitzeund Austrocknungsresistenz vin sporophyten einheimischer Polypodiaceen. Flora 155: 123-166. KIELLAND, K. 1994. Amino acid absorption by artic plants: implications for plant nutrition and nitrogen cycling. Ecology 75: 2373-2383. ______. 1997. Role of free amino acids in the nitrogen economy of artic cryptogams. Ecoscience 4: 75-79. KLEKOWSKI, E. J. J. 1969. Reproductive biology of th e pteridophyta. II. Theoretical considerations. Botanical Journal of the Linnean Society 63: 347-359. KLOCKARS, A. J., and G. SAX. 1986. Multiple comparisons. Sage Publications., Thousand Oaks. KNOWLES, R., and T. H. BLACKBURN. 1993. Nitrogen isotope techniques. Academic Press, New York. KOLB, K. J., and R. D. EVANS. 2003. Influence of nitrogen source and concentration on nitrogen isotopic discrimination in two ba rley genotypes (Hordeum vulgare L.). Plant Cell And Environment 26: 1431-1440. LIPSON, D., and T. NASHOLM. 2001. The unexpected versatil ity of plants: organic nitrogen use and availability in terrestrial ecosystems. Oecologia 128: 305-316. MILLER, J. H. 1968. Fern gametophytes as experimental material. Botanical Review 34: 361-&. MOTTIER, D. M. 1927. The behavior of certain fern prothallia under prolonged cultivation. Botanical Gazette 83: 244-266. MULKEY, S. S., and R. W. PEARCY. 1992. Interactions between acclimation and photoinhibition of photosynthesis of a tropical forest understory herb, Alocasia macrorrhiza during simulated canopy gap formation. Funct. Ecol. 6: 719-729. NADELHOFFER, K. F., AND B. FRY. 1988. Controls On Natural N-15 And C-13 Abundances In Forest Soil Organic-Matter. Soil Science Society Of America Journal 52: 1633-1640.

PAGE 93

84 NADELHOFFER, K. J., and B. FRY. 1994. Nitrogen isotope studie s in forest ecosystems. In K. Lajtha and R. H. Michener [eds .], Stable Isotopes in ecology and environmental science, 22-44. Blackwe ll scientific publi cations, Oxford. NADKARNI, N. M., and T. MATELSON. 1991. Litter dynamics within the canopy of a neotropical cloud forest, Monteverde, Costa Rica. Ecology 72: 849-860. NAYAR, B. K., and S. KAUR. 1971. Gametophytes of homosporous ferns. The Botanical Review 37: 295-396. NIKLAS, K. 1997. The Evolutionary Biology of Plan ts. University of Chicago, Chicage. NOBEL, P. S. 1978. Microhabitat, Water Relations, a nd Photosynthesis of a Desert Fern, Notholaena-Parryi. Oecologia 31: 293-309. OLIVER, M. J., B. D. MISHLER, and J. E. QUISENBERRY. 1993. Comparative Measures of Desiccation-Tolerance in the Tortula-Ru ralis Complex.1. Variation in Damage Control and Repair. American Journal of Botany 80: 127-136. OLIVER, M. J., Z. TUBA, and B. D. MISHLER. 2000. The evolution of vegetative desiccation tolerance in land plants. Plant Ecology 151: 85-100. ONG, B. L., and M. L. NG. 1998. Regeneration of drought-stressed gametophytes of the epiphytic fern, Pyrrosia p ilosellodes (L.) Price. Plant cell reports 18: 225-228. PECK, J. H. 1980. Life history and re productive biology of the fe rns of Woodman Hollow, Webster County, Iowa. Ph.D. Dissertat ion, Iowa State University, Ames. PECK, J. H., C. J. PECK, and D. R. FARRAR. 1990. Influences of lif e history attributes on formation of local and distant fern populations. American Fern Journal 80: 126142. PICKETT, F. L. 1913. Resistance of the prothallia of Camptosorus rhizophyllus to desiccation. Bul. Torrey Bot. Club 40: 641-645. ______. 1914. Some ecological adaptations of certain fern prothallia Camptosorus rhizophyllus Link., Asplenium platyneuron Oakes. American Journal of Botany 1: 477-498. ______. 1931. Notes on xerophytic ferns. America Fern Journal 21: 49-57. PILLAI, R. S., and B. L. ONG. 1999. Effects of inorganic ni trogen availability on the sporophytes of Acrostichum aureum L. Photosynthetica 36: 259-266.

PAGE 94

85 POREMBSKI, S., and W. BARTHLOTT. 2000. Granitic and gneissic outcrops (inselbergs) as centers of diversity for desicca tion-tolerant vascular plants. Plant Ecology 151: 19-28. PRANGE, R. K., and D. P. ORMROD. 1982. Effects Of Ammonium And Nitrate Nutrition On The Ostrich Fern (Matteuccia-Struthiopteris). Canadian Journal Of Plant Science 62: 195-201. PROCTOR, M. 2001. Patterns of desiccation tole rance and recovery in bryophytes. Plant Growth Regulation 35: 147-156. PROCTOR, M. C. F. 2000. The bryophyte paradox: toleranc e of desiccation, evasion of drought. Plant Ecology 151: 41-49. ______. 2003. Comparative ecophysiological measur ements on the light responses, water relations and desicc ation tolerance of the filmy ferns Hymenophyllum wilsonii Hook. and H-tunbrigense (L.) Smith. Annals of Botany 91: 717-727. PROCTOR, M. C. F., and V. PENCE. 2002. Vegetative tissu es: Bryophytes, vascular resurrection plants a nd vegative propagules. In M. Black and H. W. Prichard [eds.], Desiccation and survival in plants: drying without dying, 207-237. CAB International, Wallingford, UK. PRYER, K. M., A. R. SMITH, and J. E. SKOG. 1995. Phylogenetic rela tionships of extant ferns based on evidence from morphology and rbcL sequences. American Fern Journal 85: 205-282. PUTZ, F. E., and N. M. HOLBROOK. 1989. Strangler fig root ing habits and nutrient relations in the llanos of Venezuela. American Journal of Botany 76: 781-788. RACUSEN, R. H. 2002. Early development in fern gametophytes: Interpreting the transition to prothallial architecture in terms of coordinated photosynthate production and osmotic ion uptake. Annals of Botany 89: 227-240. REICH, A., J. J. EWEL, N. M. NADKARNI, T. DAWSON, and R. D. EVANS. 2003. Nitrogen isotope ratios shift with plan t size in tropical bromeliads. Oecologia 137: 587590. ROBINSON, D. 2001. delta N-15 as an integrator of the nitrogen cycle. Trends In Ecology & Evolution 16: 153-162. ROBINSON, S. A., J. WASLEY, M. POPP, and C. E. LOVELOCK. 2000. Desiccation tolerance of three moss species from continental Antarctica. Australian Journal of Plant Physiology 27: 379-388.

PAGE 95

86 SANFORD, R. L., P. PAABY, J. C. LUVALL, and E. PHILLIPS. 1994. Climate, geomorphology, and aquatic systems. In L. A. McDade, K. S. Bawa, H. A. Hespenheide, and G. S. Hartshorn [eds.] La Selva: Ecology and Natural History of a Neotropical Rain Forest, 19-33. Un iversity of Chicago Press, Chicago. SAS-INSTITUTE. 2005. JMP User's Guide. SA S Instite, Inc., Cary. SATO, T., and A. SAKAI. 1980. Freezing resistance of gamet ophytes of the temperate fern, Polystichum retroso-paleaceum. Canadian journal of botany 58: 1144-1148 ill. ______. 1981. Cold tolerance of gametophytes of some cool temperare ferns native to Hokkaido. Can. J. Bot. 59: 604-608. SCHEINER, S. M., and J. GUREVITCH. 2001. Design and analysis of ecological experiments. Oxford University Press, Oxford. SCHMIDT, A., and G. R. STEWART. 2003. Delta 15N values of tropical savanna and monsoon forest species reflect root sp ecilizations and soil nitrogen status. Oecolgia 134: 569-577. SCHMIDT, G., S. STUNTZ, and G. ZOTZ. 2001. Plant size: an ignored parameter in epiphyte ecophysiology? Plant Ecology 153: 65-72. SCHMIDT, S., and G. R. STEWART. 1999. Glycine metabolism by plant roots and its occurrence in Australian plant communities. Australian Journal Of Plant Physiology 26: 253-264. SMITH, D. L. 1972a. Localization of phosphatases in young gametophytes of Polypodium vulgare L. Protoplasma 74: 133-148. ______. 1972b. Staining and osmotic properties of young gametophytes of Polypodium vulgare L. and their bearing on rhizoid function. Protoplasma 74: 465. TRESEDER, K. K., and P. M. VITOUSEK. 2001. Effects of soil nut rient availability on investment in acquisition of N and P in Hawaiian rain forests. Ecology 82: 946954. TUOMISTO, H., and K. RUOKOLAINEN. 1994a. Distribution of Pteridophyta and Melastomataceae Along an Edaphic Gradie nt in an Amazonian Rain Forest. Journal of Vegetation Science 5: 25-34. ______. 1994b. Distribution of Pteridophyta a nd Melastomataceae Along an Edaphic Gradient in an Amazonian Rain-Forest. Journal of Vegetation Science 5: 25-34. TUOMISTO, H., and A. D. POULSEN. 1996. Influence of edaphic specialization on pteridophyte distribution in ne otropical rain forests. Journal of biogeography 23: 283-293.

PAGE 96

87 TUOMISTO, H., and A. DALBERG. 1996. Influence of edaphic specialization on pteridophyte distributions in neotropical rain forests. Journal of Biogeography 23: 283-293. TUOMISTO, H., and A. D. POULSEN. 2000. Pteridophyte diversity and species composition in four Amazonian rain forests. Journal of vegetation sc ience: official organ of the International Association for Vegetation Science 11: 383-396. TUOMISTO, H., A. D. POULSEN, and R. C. MORAN. 1998. Edaphic distribution of some species of the fern genus Adiantum in western Amazonia. Biotropica 30: 392399. WALKER, L. R., and G. H. APLET. 1994. Growth And Ferti lization Responses Of Hawaiian Tree Ferns. Biotropica 26: 378-383. WALP, R. L. 1951. Fern Prothallia unde r Cultivation for 12 Years. Science 113: 128-129. WANEK, W., S. K. ARNDT, W. HUBER, and M. POPP. 2002. Nitrogen nutrition during ontogeny of hemiepiphytic Clusia species. Functional Plant Biology 29: 733-740. WATANABE, M., T. KIKAWADA, N. MINAGAWA, F. YUKUHIRO, and T. OKUDA. 2002. Mechanism allowing an insect to surv ive complete dehydration and extreme temperatures. Journal Of Experimental Biology 205: 2799-2802. WATKINS, J. E., and D. R. FARRAR. 2005. Origin and taxonomic affinities of Thelypteris (subgen. Stegnogramma) burksiorum (Thelypteridaceae). Brittonia 57: 183-201. WATKINS, J. E., C. CARDELUS, R. K. COLWELL, AND R. C. MORAN. 2006. Species richness and distribution of ferns along an elevational gradient in Costa Rica. American Journal of Botany 93: 73-83. WATKINS JR., J. E., and D. R. FARRAR. 2005. Origin and taxonomic affinities Thelypteris (subgen. Stegnogramma) burksiorum (Thelypteridaceae). Brittonia 57: 183-201. ZOTZ, G. 1997. Photosynthetic capacity increases with plant size. Botanica Acta 110: 306-308. ZOTZ, G., A. ENSLIN, W. HARTUNG, and H. ZIEGLER. 2004. Physiological and anatomical changes during the early ontogeny of th e heteroblastic bromeliad, Vriesea sanguinolenta, do not concur with the mor phological change from atmospheric to tank form. Plant Cell And Environment 27: 1341-1350.

PAGE 97

88 BIOGRAPHICAL SKETCH James Edward (Eddie) Watkins, Jr. wa s born on 12 March, 1974 in Ozark, Dale County, Alabama. He attended the now-conde mned Flowers Elementary School where his first introduction to scienc e was a project on turtles by hi s first grade teacher Mrs. Hopper. He graduated to attend D.A. Smith Mi ddle School. It was th ere, under direction of his eighth grade teacher, Dena Byers he first began to develop an understanding of scientific experiment. He competed in severa l regional science fairs, making it as far at the Alabama State Science Fair for his work on factors controlling the rate at which mice could exit a complicated maze. During these years, he spent much of his time fishing, hunting, building forts, and observing nature from his daily hikes in to the forests surrounding his home. During these early years he developed a tr ue connection with the natura l world. He gained his early understanding of how this world was put t ogether by his first biological mentor Ms. Linda Dees: science teacher at Carroll Hi gh School. During these formative high school years he began to put together what he would do for the rest of his life. One of the most important developments came when Ms. Dees required his freshman biology class to complete a plant collection. This Eddie did by only collecting live ferns that were then transplanted into the schools nature preserve. In th e course of this collection, he discovered two of the rarest ferns in Alabam a and went on to publish some of this work in a peer-reviewed journal. During his early fern forays, he made the acquaintance of Professor Warren Herb Wagner, Jr.; the world s leading fern authority at the time, and

PAGE 98

89 continues to inspire Eddies work toda y. After graduation, he attended Auburn University, where he worked under the direc tion of Dr. John D. Freeman and Dr. Robert S. Boyd. In the lab and courses of Dr. Boyd that Eddie was finally able to put nature and experimental sciences together and begin to develop an understanding of the complexities of ecology. After graduation, he attended Iowa State University under Dr. Donald Farrar where he attained an MS degree with his thesis on Thelypteris burksiorum These years were paramount to his Pteridological developm ent, as Dr. Farrar is one of the last oldschool fern biologists and was as smitten with ferns as Eddie. Eddie graduated in 2000 and spent a year living with his wife Cath erine in Costa Rica, studying the magnificent array of ferns at La Selva Biological Stati on and beyond. He then returned to the South where he began his doctoral studies with Dr. Stephen Mulkey and Dr. Michelle C. Mack. After completing of his doctoral studies, E ddie will begin a post doctoral fellowship in the lab of Michelle Holbrook at Harvard University.


Permanent Link: http://ufdc.ufl.edu/UFE0013484/00001

Material Information

Title: Functional Ecology of the Gametophytes and Sporophytes of Tropical Ferns
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013484:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013484/00001

Material Information

Title: Functional Ecology of the Gametophytes and Sporophytes of Tropical Ferns
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013484:00001


This item has the following downloads:


Full Text









FUNCTIONAL ECOLOGY OF THE GAMETOPHYTES AND SPOROPHTYES
OF TROPICAL FERNS















By

JAMES EDWARD WATKINS, JR.


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006




















To my parents James E. Watkins and Juanita P. Watkins for encouraging my curiosity in
the natural world, for spending hours looking for material to add to my annual
elementary-school leaf collections, and for writing sick notes to my high school principal
so that I could skip class and spend the early days of Spring in search of the elusive
Botrychium lunarioides















ACKNOWLEDGMENTS

Many people and organizations contributed to the successful completion of my

work. I thank my supervisory committee members (Stephen Mulkey, Michelle Mack,

Thomas Sinclair, and Pamela Soltis) for their many contributions and their patience

throughout. I also thank members of the University of Florida Ecology Group

(specifically Louis Santiago, Juan Posada, Grace Crummer, Jordan Major and Jason

Vogel) for all of their help. I am especially indebted to Jason Vogel for his countless

hours of statistical discussions and for his ability to raze my radical political ambitions.

Throughout my doctoral studies, I often relied on the sage advice of Dr. Jack Ewel be it

academic, personal, or hunting: I am grateful. I also thank Robbin Moran (New York

Botanical Garden), and Donald Farrar (Iowa State University, Ames) for always lending

an ear or helping hand when it was needed most.

Some of this work took place at La Selva Biological Station in Costa Rica and I am

indebted to the Organization for Tropical Studies for opening the doors to the tropical

world to me. I also thank my wife, Catherine Cardelus, for her immense patience and

support throughout this process. I also thank my son Santiago for teaching me what life is

really all about. Funding was provided by the National Science Foundation, the

Organization for Tropical Studies, the Mellon Foundation, the American Fern Society,

and the University of Florida Graduate School and Department of Botany.
















TABLE OF CONTENTS



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

LIST OF TABLES .............. ................................................. ....... vi

L IST O F F IG U R E S .... ......................................................... .. .......... .............. vii

ABSTRACT .............. ......................................... ix

CHAPTER

1 IN T R O D U C T IO N ............................................................................. .............. ...

2 GAMETOPHYTE ECOLOGY AND DEMOGRAPHY OF TROPICAL
EPIPHYTIC AND TERRESTRIAL FERNS ............................................................5

In tro du ctio n ....................................................................................................... .... .. 5
M materials and M methods ................................................................. ........................ 6
S tu d y S ite ................................................... 6
G am etophyte Transects ............................................................................. 7
D istu rb an ce P lots ............................................................................... 7
D em ography ....................................... ..............
G am etophyte Survival A analysis ................................... ....................................... 9
R e su lts ...................................... .......................................................10
T ran sects ...................................... ............................... ................ 10
Disturbance Plots ...................................................................... ... ...... 11
D em og rap hy ................................................................12
D isc u ssio n ............................................................................................................. 1 4
Gametophyte Distributions........................................................ 14
D ensity and Species R ichness ........................................ ......... ............... 17
Sporophyte E cology ............................................ .. ........ .............. 18
C o n c lu sio n s ........................................................................................................... 1 9

3 COMPARATIVE DESICCATION TOLERANCE OF TROPICAL FERN
GAMETOPHYTES: ECOLOGICAL AND EVOLUTIONARY
C O N SE Q U E N C E S .......................................................................... .....................2 9

Introdu action ...................................... ................................................. 2 9
M materials and M methods ....................................................................... ..................33









Spore Material and Growth Conditions...........................................................33
Desiccation Experiments ........... ......... .......................................... 33
Chlorophyll-Fluorescence M easurements.................... .............................. ..35
Statistical A analysis .......................... .......... ............... .... ..... .. 35
Results ................ ................................ ...............36
D esiccation Survey ....... .. ..... ..... .. ...... .. .. ......... .... ... ............ ... 36
D esiccation R ates ........................................ .. .. .... ........ ..... .... 37
D esiccation C ycles ........................................ .................. ........ 38
D discussion ......... ..... .... .......................................................... ......38
V aviation in VPD ............ .. ............................... ........ .... .. ............ 39
D esiccation C ycles ........................................ .................. ........ 41
C o n c lu sio n s........................................................................................................... 4 3

4 NITROGEN-15 NATURAL ABUNDANCE AND NITROGEN USE
STRATEGIES OF THE GAMETOPHYTES AND SPOROPHYTES OF
TROPICAL EPIPHYTIC AND TERRESTRIAL FERNS........................................55

Intro du action ...................................... ................... ............................ 5 5
M material and M methods .......................................................... .. ............... 57
Study Site............................................. 57
Study Species......................................................................57
Isotopic Natural Abundance and 615N Labeled Uptake ....................................59
N utrient U ptake Calculations ........................................ ......................... 60
R e su lts................... ........... ........ ........... .............. ......... ................ 6 0
615N Natural Abundance and N concentration (mg g) ............................... 60
615N L abeled U ptake ............................................ .. .. .. ...... ........... 6 1
D isc u ssio n ...................... .. ............. .. ......................................................6 2
C o n c lu sio n s........................................................................................................... 6 7

5 CON CLU SION S ................................................................75

LIST OF REFERENCES ... ................................ .......... ..............................79

B IO G R A PH ICA L SK ETCH .......................................................................... ... 88


















v















LIST OF TABLES


Table page

2-1 Relationship of gametophyte density and richness with three levels of
experimental disturbance and two light levels .............................. ...................21

2-2 Demographic and survival analyses for the gametophytes of 5 fern species using
the Wilcoxon test to compare survival distribution functions for different species.22

3-1 Species and life form from the initial desiccation survey ........................................44

3-2 Fv/Fm recovery results from the repeated measures ANOVA for gametophytes
exposed to three different desiccation intensities................... .................45

3-3 Fv/Fm recovery results from the repeated measures ANOVA for gametophytes
exposed to 1, 2, or 3 desiccation cycles ....................................... ............... 46

4-1 Species, life form and ecology for the natural abundance and uptake experiments 68















LIST OF FIGURES


Figure page

2-1 Number of gametophytes counted and their relation to disturbance from natural
transects ........... ................................... ......... .................... 23

2-2 The percentage of fern gametophytes as influenced by type of disturbance............24

2-3 The relationship between canopy openness and gametophyte density for A.
canopy and B. terrestrial species ........._. ..... ...... .......... ........... ... ............ 25

2-4 Gametophyte densities as influenced by light and disturbance in experimental
p lo ts. ............................................................. ................ 2 6

2-5 Mean longevity (months) A) and percent gametophytes still alive and un-
recruited B) for the 25-month period of the study .................................................27

2-6 Kaplan-Meier survivorship curves A) and proportion recruiting B) of 5 species
of fern gametophytes over the 25-month study period. No data were collected
during m months 4-7 and 15-24 ............................................................................28

3-1 A) Gametophyte drying curves from 12 tropical fern species of different
habitats. Species were exposed to a VPD to 1.32KPa (-50%RH) for 45 min. B)
Depression of photochemical efficiency in the same gametophytes over a series
of decreasing thallus w ater contents ....................................... ............ ............... 48

3-2 The rate of absolute water loss relative to gametophyte size as indexed by dry
m ass ...............................................................................................4 9

3-3 A) Rate of thallus drying as calculated from 3-1 for 12 tropical fern species of
different habitats. (B) Proportional recovery of the pre-treatment dark adapted
value of Fv/Fm in these same species ........ ................................................. ............... 50

3-4 Proportional recovery of the pre-treatment dark adapted value of Fv/Fm and rate
of thallus water loss expressed as A) relative water content ((g fresh weight- g
dry weight)/g saturated weight g dry weight))* 100 and B) absolute water
content (g w et w eight/g dry w eight) ............................................. ............... 51









3-5 Fv/Fm recovery graphs for gametophytes of A) Diplazium subsilvaticum, B)
Phlebodium pseudoaureum, (c) Polypodium triserale exposed to three different
desiccation intensities: VPD-0.53kPa (20%RH), VPD-1.32kPa (50%RH), and
V P D 2 .12kP a (80% R H ) .............................................................. .....................52

3-6 Proportional Fv/Fm recovery results for gametophytes exposed to 1, 2, or 3
desiccation cycles at VPD- 1.32kPa (50%RH). .............................................. 53

3-7 Morphology in fern gametophytes is diverse and is closely related to species
ecology and phylogeny................................................... .............................. 54

4-1 Sporophtyic and Gametophytic 615N natural abundance signatures of 10 tropical
fern species. ...........................................................................69

4-2 Sporophtyic and Gametophytic (a) 615N natural abundance signatures and (b) N
concentration (mg g-1) of epiphytic, terrestrial, and hemiepiphytic tropical fern
sp e cie s. ........................................................................... 7 0

4-3 615N natural abundance signatures from the hemiepiphytic fern Lomariopsis
v e s tita ............................................................................ 7 1

4-4 Uptake curves from 615N labeled solutions................................... ............... 72

4-5 Uptake saturation values (Vmax) of each N form derived from Michaelis-Menten
functions of the data from Fig. 4-2................................... ...................... .. .......... 73

4-6 12 Uptake saturation values (Km) of each N form derived from Michaelis-
M enten functions of the data from Fig. 4-2 ............. ...............................................74















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

FUNCTIONAL ECOLOGY OF THE GAMETOPHYTES AND SPOROPHTYES OF
TROPICAL FERNS

By

James E. Watkins, Jr.

May 2006

Chair: Stephen S. Mulkey
Cochair: Michelle Mack
Major Department: Botany

Ferns are an important part of both temperate and terrestrial floras, yet their

ecology remains poorly understood. Although ferns are dispersed by tiny wind-blown

spores, most species are limited to specific habitats; on local levels, ferns are no more

widespread than angiosperms. One aspect of fern biology that poses unique ecological

problems is the dependence on a free-living gametophyte. I examined the autecology and

ecophysiology of the fern gametophyte to understand this structure's role in shaping fern

distributions. My study showed that the gametophytes of epiphytic and terrestrial ferns

respond differently to light, disturbance, and desiccation stress, and show unexpected

versatility in nutrient relations. In all cases, such variation is closely linked to species

ecology. Selective pressures acting on the gametophyte generation may be largely

responsible for species distributions.














CHAPTER 1
INTRODUCTION

The ferns, with some 12,000 species, are the third most species group of land

plants following angiosperms and bryophytes. Their intermediate evolutionary position

affords the group a combination of both non-vascular and vascular plant life histories.

Ferns rely on a supposedly delicate, short-lived, and usually haploid independent

gametophyte (one of their connections to non-vascular plants), and in some, recruitment

from the gametophyte often follows into the usually diploid sporophyte stage with

lignified vascular tissue (their most obvious connection to vascular plants). It is the

ecology of the gametophyte in this choreographed alternation of generations that remains

largely unstudied. Such scientific deficit has been commented upon for decades (Pickett,

1914; Holttum, 1938; Cousens, Lacey, and Kelly, 1985; Greer and McCarthy, 1999), yet

there has been little movement to increase our understanding of basic fern ecology.

Richard Eric Holttum (1895-1990) is arguably the founding father of fern ecology.

Professor Holttum was trained at Cambridge and spent much of his time teaching and

studying the ferns of Southeast Asia. By his hand one of the first great treatises of fern

ecology was written (Holttum 1938). His seminal "The ecology of tropical pteridophytes"

first approached the topic in a synthetic way by incorporating intimate knowledge of

ferns and combining it with careful observation and questioning to get to a big-picture

conclusion about the ecology of the group. He also addressed ecology of the gametophyte

generation:









We are accustomed to see and to marvel at the great varied form and adaptation of
the sporophtyes, which are the ferns as we know them, but indeed there must be
nearly as much variety of adaptation among the gametophytes. It is true that if the
prothallus ofPlatycerium grew upon the forest floor, the resulting sporophyte, if
produced, would find itself in uncongenial surroundings, and would not develop
very far; but it is also true that the Platycerium prothallus must be able to develop
in relatively exposed position on the tree trunk in which prothalli of many ferns
would be unable to exist (Holttum 1938, pp. 421-422).

One of Holttum's key observations was the recognition of gametophyte-mediated

controls on fern recruitment. Platycerium, commonly know as staghorn ferns, are Old

World epiphytes with diverse ecology; many species grow on highly exposed emergent

canopy tree trunks, but never on the forest floor. This observation combined with the

copious production of wind-dispersed spores showed that dispersal is perhaps of only

modest importance in the distribution of ferns: the gametophyte played the critical role in

recruitment. Unfortunately, Holttum's call to arms was largely ignored and we have

progressed little since the publication of his work.

Many factors have limited the study of fern gametophyte ecology. Avoidance of

the gametophyte generation may have been driven by some of the comments made by

Frederick Orpen Bower (Bower, 1923) in "The Ferns". Bower saw little taxonomic value

in fern gametophytes and doubted their utility in advancing pteridology. Bower was such

a recognized figure (and the depauperate literature of the time basically supported his

ideas) that many botanists took his words to heart. Fortunately, many studies have since

incorporated gametophytic characters into phylogenies, and we have learned that

gametophytes have systematic value and have many taxonomic characters that allow for

species or morpho-type identification (Atkinson and Stokey, 1964; Nayar and Kaur,

1971; Chiou and Farrar, 1997; Watkins Jr. and Farrar, 2005).









There has been limited application of modem ecological methodology and

statistics to the study of fern gametophyte ecology. In a series of observational field and

elegant lab experiments, Pickett (1913; 1914) showed that the gametophytes of

Asplenium rhizophyllus and A. platyneuron could be long-lived and survive winter

temperatures and extreme drought. Pickett's work helped usher in a new way of thinking

about fern gametophyte ecology and later reports by Mottier (1927) and Walp (1951)

showed that the gametophytes of some temperate species were essentially indeterminate

and could grow in vitro for decades, if reproduction were prevented. This notion was

further supported by Donald Farrar who developed the now classic story of tropical

gametophytes growing independently of sporophytes in the temperate Appalachian

Mountains (Farrar, 1967, 1971; Farrar, 1998). In many cases, these species form

populations of asexually reproducing gametophytes that number in the tens of thousands

and appear to survive winter freezes and summer droughts.

In a series of papers, Michael Cousens developed the concept of gametophyte safe

sites and showed that multiple factors act at the level of the gametophyte to shape their

distribution and recruitment (Cousens, 1979, 1981; Cousens, Lacey, and Kelly, 1985;

Cousens, 1988; Cousens, Lacey, and Scheller, 1988). Cousens' work largely developed in

the backdrop of earlier studies showing that fern gametophytes have a distinct ecology

relative to sporophytes.

From the turn of the century with Pickett's work, to more recent work by Cousens

and Farrar, we have learned that gametophytes in temperate forests have a distinct

ecology relative to sporophytes, that gametophytes can be long-lived in natural and

especially in vitro settings, and robust when dealing with abiotic stresses. Yet, the






4


ecology of tropical gametophytes remains essentially unstudied. The goal of this

dissertation is to examine multiple aspects of fern ecology, focusing on the gametophyte

generation.














CHAPTER 2
GAMETOPHYTE ECOLOGY AND DEMOGRAPHY OF TROPICAL EPIPHYTIC
AND TERRESTRIAL FERNS

Introduction

Ferns are conspicuous components of temperate and especially tropical wet

forests. Yet, general fern ecology is poorly understood. Much early work was anecdotal

or derived from studies and observations made from sporophytes or comments obscured

in floristic inventories. A flurry of recent studies attempted to describe both the patterns

of fern sporophyte diversity and the causal relations behind such patterns (Tuomisto and

Ruokolainen, 1994b; Tuomisto and Dalberg, 1996; Tuomisto, Poulsen, and Moran, 1998;

Tuomisto and Poulsen, 2000; Jones et al., 2006; Watkins et al., 2006). These studies were

critical in developing ecological models to better understand the biology of the fern

sporophyte. Yet, focusing on sporophyte ecology, only told us a small part of fern

ecology. Missing are studies on the ecology of the free-living gametophyte.

Ferns alternate between two independent generations: the haploid gametophyte

and the diploid sporophyte. The gametophyte is a fundamentally different organism than

the sporophyte. It lacks vascular tissue, produces rhizoids instead of true roots, has poorly

developed to non-existent cuticles, and is comparatively small. We know from early work

that gametophytes can be more widespread and can grow in areas that are uninhabitable

to sporophtyes. Yet, recruitment happens in the gametophyte generation, and the resulting

sporophyte distributions depend on gametophyte ecology.









Gametophyte biology is complex, and ontogeny and morphology vary

tremendously among species (Atkinson and Stokey, 1964; Nayar and Kaur, 1971). A

common observation is that there are the apparent fundamental differences in

morphology and potentially longevity between epiphytic and terrestrial species (Dassler

and Farrar, 1997, 2001). Epiphytic species often produce gametophytes with diverse

morphologies that are frequently capable of asexual reproduction and are potentially

long-lived. Most terrestrial species are thought to produce the short-lived, textbook

cordate thallus and exhibit little ability to reproduce asexually (but see Watkins and

Farrar 2005). Little quantitative data have been generated to back up such longevity

claims, and I have been unable to find a single paper that describes factors that influence

the distribution and mortality of tropical gametophtyes.

The goal of our study was to examine the causal mechanisms of the distribution of

fern gametophytes and the demography of several tropical epiphytic, hemiepiphytic, and

terrestrial species. We examined the distribution of epiphytic and terrestrial gametophytes

and ask what in situ factors control gametophyte establishment. Then we examined the

gametophyte demography of 5 species from different habitat types, to assess gametophyte

survival and recruitment rates and relate these to life history.

Materials and Methods

Study Site

This study was conducted at La Selva Biological Station (Heredia Province) in the

Atlantic lowlands of northeastern in, Costa Rica. La Selva is a 1400 ha tropical wet forest

having a mean annual rainfall of about 4,300 mm, with peaks of precipitation in June-

July and November-December, and a drier period in March. Mean monthly rainfall









nevertheless never falls below 150 mm in any month during the dry season based on

long-term meteorological records.

Gametophyte Transects

In order to describe the occurrence of gametophytes in nature, 425 plots of

25 cm x 25 cm were placed along 50 randomly chosen terrestrial 50m transects, and 425

25 cm x 25 cm canopy plots were placed in 9 canopy trees. The total number of

gametophytes (irrespective of identification) was counted and recorded. Each quadrat

was coded for level of disturbance: 0=undisturbed (<5cm2 bare substrate); 1 = low

disturbance (i.e. >5cm2 bare substrate); 2 = medium disturbance (>5cm2 bare substrate

and substrate disturbed); 3= high disturbance (100% bare substrate and substrate turned

over). Additionally, each quadrat with a disturbance rating of Level 1 and above was

coded for the type of damage when possible. To assess light environment, a digital

hemispherical photograph was taken with a Nikon Coolpix 950 digital camera (Melville,

NY, U.S.A.) with a fisheye lens attachment, then analyzed using Gap Light Analyzer

software (Frazer et al., 1999) to estimate the percentage of total light transmittance.

Photos were taken 25 cm above each quadrat. To determine the influence of light

environment on density, both number of gametophytes and percent canopy transmittance

were log transformed and analyzed by regression analysis. We used ANOVA to

determine the influence of level of disturbance on gametophyte density. Unless otherwise

stated, all analyses were performed with the computer program JMP version 5.01 (SAS-

Institute, 2005).

Disturbance Plots

To better understand the influence of disturbance and light on terrestrial fern

establishment, 20 disturbance plots were established and monitored for gametophyte









density at 5 months post establishment. All plots were established on the same soil type

in primary forest, with 10 plots placed in low-light understory habitats and 10 placed in

high-light canopy gaps of similar age. Each plot measured 1m2 and was divided into for

0.5 m x 0.5 m subplots of increasing disturbance that were similar in degree to those

found in nature. The undisturbed treatment subplot acted as the control, and no leaf litter

was removed. For low disturbance level, we removed all leaf litter with no mechanical

damage to the soil. The medium disturbance level was raked with a metal sand rake to

disturb the first 5 cm of soil. The high disturbance level was physically turned over with a

shovel, to a depth of approximately 20 cm. Gametophyte density and diversity were

recorded in the center 25 cm2 area. Litter-fall was removed from the disturbed plots

weekly, and after 5 months, plots were assessed for density (and when possible, diversity

of gametophytes). Light environment was determined with digital photography as

discussed above. Determination of gametophyte identity was difficult, and individuals

were thus lumped into "types" that in actuality may represent multiple species. Types

were assigned based on morphological characters that were identifiable by the use of a

10X or 20X hand lens and were identified and organized based on: trichome presence and

type, rhizoid color, gametophyte shape, and the presence and morphology of gemmae.

Identification of fern species from gametophytic characters was complicated and should

be taken as a conservative estimate of actual species richness. Only those gametophytes

that were mature were counted. A 2X4 full factorial ANOVA was used to determine the

effects of both light and disturbance intensity on gametophyte density and diversity.

Demography

In June of 2003, gametophytes from three populations of each of 5 species (See

table 2-2) were located and marked in the field. Marked individuals were checked once









each month and followed for the next 15 months with one final census made at 25

months. No data were recorded during the 5-7th month. At each census, individuals were

recorded as present, dead or missing, or as recruits into the sporophyte generation. When

possible, individuals were coded for their cause of mortality.

In the case of terrestrial species, gametophytes were marked with a numbered

aluminum nail; whereas, the epiphytic species were either marked with a nail or with a

numbered tag attached to the substrate with copper wire. It was not possible in all cases

to determine precisely the initial age of marked gametophytes. Therefore, individuals

were chosen according to their initial size. Initial sizes were held constant within a

species but differed among the species. Longevities were calculated as the time between

the initial mark (treated as birth) and death of each gametophyte. Species were chosen to

represent different functional types as discussed below. A major flood event took place in

month 11; individuals were sampled three days before the flood (for the regularly

scheduled 11 month survey) and then three days after the flood to serve as an extra

survey to determine the influence of flooding. The next sample period took place on the

next corresponding survey day and was recorded as the month 12 survey period. This

allowed for more precise determination of mortality due to flooding rather than

categorizing these individuals in the unknown category.

Gametophyte Survival Analysis

As with many demographic studies, individuals can leave the study by different

avenues. Such absent samples were coded as right-censored data points (Hollander and

Wolf, 1999). In this study, only those individuals that recruited into the sporophyte

generation and those still alive at the end of the experiment (25 months) were recorded as

censors. Censoring individuals reduces the sample size of individuals at risk after the









time of censorship. Censoring, therefore, reduces the number of individuals contributing

to the curve, and each death after a censored point represents a higher proportion of the

remaining population. Subsequent deaths will result in greater decreases in overall

survivorship. Censorships that occur early in the study have a greater effect on

survivorship curves than those removed at later periods. Thus, the data from the

survivorship curve after the first censor represent an estimate and not the actual

survivorship of the population. In order to clarify the survival curves, we also plotted the

cumulative proportion of individuals that recruited at each time interval.

Gametophyte survival functions were estimated using non-parametric Kaplan-

Meier product-limit survival functions (Collett, 2003). These analyses were also used to

estimate mean life span for each species. Log-rank X2 statistics were computed to test for

homogeneity of the survival functions for all species. Weibull distributions were used to

model survivorship functions and to calculate the parameters a and P. The scale

parameter a is a measure of the degree of hazard for the species; whereas, the shape

parameter 0 determines the degree of change in the hazard function over time. Large

values of a correspond to low hazard levels (i.e. greater survivorship) where low values

equate to rapidly decaying survivorship. Large values of 0 (i.e. >1) correspond to an

increasing hazard rate that affects older individuals over younger individuals. With a P <1

younger individuals are more likely to die within the period of the experiment.

Results

Transects

A combined total of 2096 gametophytes were sampled with 329 recorded from

canopy, 538 from low-trunk, and 1229 from terrestrial habitats. Level of disturbance had

a highly significant effect on the number of gametophytes in terrestrial habitats (Fig. 2-









2b, r2=0.65, F=53.58, p<0.001) with greater percentages of gametophytes occurring in

more disturbed habitats. Less than 1% of terrestrial gametophytes were found in areas

without disturbance. The opposite trend was apparent in canopy habitats where level of

disturbance had less influence on numbers of gametophytes (Fig. 2-2a, r2=0.11, F=2.40,

p=0.08) and greater percentages of gametophytes occurred in less disturbed habitats (58%

of canopy gametophytes were found in areas with no disturbance).

In terrestrial transects, seven causes of disturbance were identified: leaf-litter

removed, new and old root tip-ups from fallen trees, rotten logs, erosion, branch falls, and

animal causes (Fig. 2-2b). A total of six causes of disturbance were identified in canopy

habitat: insect, branch falls, epi-slides, physical damage, animal, and unknown causes

(Fig. 2-2a). Identification of causes in the low-trunk habitat was difficult and thus was

excluded from all analyses. In the case of terrestrial species, recent root tip-ups harbored

the greatest number of gametophytes (>50%). The disturbance category with the greatest

percentage of gametophytes in canopy habitats was animal disturbance with -20%. In

addition to disturbance, canopy openness (as a surrogate for light level) exhibited a

positive effect on the number of terrestrial gametophytes (Fig. 2-3a, r2=0.423,

p=<0.0001), but exhibited little influence on canopy gametophyte density (Fig. 2-3b,

r2=0.001, p=<0.539).

Disturbance Plots

A total of 1247 gametophytes from 16 morpho-types were counted in the

experimentally disturbed plots. There were 6 non-unique morpho-types found in the low

light treatment. There were a total of 16 types with 10 unique types in the high light

treatments. There was a significant and positive effect of both increasing light and

disturbance and the interaction of light and disturbance on gametophyte density among









the plots (Fig. 2-4, Table. 2-1). Likewise, morpho-type richness was significantly

influenced by both light and disturbance but exhibited no significant interaction (Fig. 2-4,

Table.2-1).

Demography

All combined, 809 gametophytes from the five species were marked and followed

throughout the demography study. A total of 263 gametophytes were marked from the

understory terrestrial species Danaea wendlandii. The three populations of this species

were all recorded in the understory of primary forests from sites that were at least 50m

from trail sides. We marked 275 gametophytes of Pityogramma ebenea, an abundant

species often found in full to partial sun in disturbed sites such as road and trail sides. All

populations of this species were recorded from disturbed sites within the forests or in

open areas away from trail sides. Sixty-seven gametophytes of the understory

hemiepiphyte Lomariopsis vestita were marked from small diameter trees in primary

forests. Two canopy epiphytes were also marked: 98 from the high-light epiphyte Vittaria

lineata, and 106 from the medium-light understory epiphyte Campyloneurum

brevifolium.

Survival distribution functions varied significantly among species for the entire

survey period (Table 2-2, log-rank X2 = 386.2, d.f. = 4, p < 0.0001). Campyloneurum

brevifolium had the highest mean longevity and was 5 times that of the lowest:

Pityrogramma ebenea (Fig. 2-5a, 2-6a). When combined, the epiphytic species: C.

brevifolium, Lomariopsis vestita, and Vittaria lineata had higher mean longevities than

the terrestrial species (log-rank 2 = 212.3, d.f. = 1, P < 0.0001). In all cases 0 >1,

indicating an increasing hazard rate suggesting that older individuals are more likely to

die than younger individuals over the study period. Percent recruitment varied among the









species, with C. brevifolium < V. lineata < D. wendlandii < L. vestita ~ P. ebenea. The

cumulative proportion of gametophytes recruiting varied for all species over the sampling

time of the study (Fig. 2-6b). Initial recruitment was highest for both terrestrial species.

More than 30% of gametophytes of P. ebenea had recruited between plot establishment

and the first census. No additional individuals recruited beyond month eight. Initial

recruitment was lower for D. wendlandii, but increased throughout the study period.

Recruitment was lowest for V lineata and C. brevifolium with essentially no recruitment

occurring after the third census up until the 25th month (Fig. 2-6b). The percent of

gametophytes still alive at the end of the study also varied from 0% in P. ebenea to just

over 70% in C. brevifolium (Fig. 2-5b).

A total of 7 causes of mortality, including an unknown category, were surveyed in

the field. Catastrophic habitat failure occurred when habitats were over 95% of the

individuals were destroyed. This happened when entire trees fell in the case of epiphytes

or when hill sides collapsed with some terrestrial species. Flooding also resulted in

catastrophic failure, but was separated as unique disturbance type because we were able

to directly assess its influence (Fig 2-2a). Minor erosion also resulted in the loss of some

individuals as did a massive flood in month eight of the study. Fungal attack, herbivory,

and physical damage (as would occur from a branch or rock fall that physically removed

individuals from the population) were also identifiable causes of mortality. The unknown

category likely consisted of a contribution of all of these plus unidentifiable novel

disturbances. Each cause resulted in different magnitudes of mortality, with some

categories completely absent from some species and/or populations (Fig. 2-2). Most %









mortality was attributable to unknown causes such as an individual simply missing from

the population without any sign of disturbance, etc.

Discussion

Gametophyte Distributions

The present study clearly demonstrates the importance of disturbance for

gametophyte establishment. Even minor disturbances that remove leaf litter and turn up

the soil can produce sites for gametophyte establishment. Disturbance that physically

turns up soil not only produces an exposed and competition free habitat, is can also

exposes the underlying spore bank and provide additional propagules that may further

contribute to density and richness. Surprisingly, the maximum number of terrestrial

gametophytes found in the lowest level of natural disturbance was three and the majority

of the undisturbed sites had no established gametophytes.

To my knowledge, disturbance has never been reported to be an important factor

influencing gametophyte density or shaping species distributions. However, studies on

the gametophytes of temperate species have highlighted the importance of nutritional and

edaphic safe sites for the gametophytes ofLorinseria (Woodwardia) areolata (Cousens,

Lacey, and Scheller, 1988) and Blechnum spicant (Cousens, 1981). Little is known of

other factors influencing gametophyte distributions and a few temperate studies have

produced mixed results suggesting that gametophyte gender expression may influence

gametophyte distribution (Klekowski, 1969; Crist and Farrar, 1983) while others have

found no relationship (Holbrook-Walker and Lloyd., 1973; Greer and McCarthy, 1999).

Apart from these studies, little is known of the influence of these characters on tropical

fern gametophyte distributions. Numerous studies have however, examined factors

behind sporophyte distributions and have fingered important roles of microclimate









(Nobel, 1978) and water availability in dry sites (Marquez et al. 1997) and edaphic

factors (Tuomisto and Ruokolainen, 1994a; Tuomisto and Poulsen, 1996; Tuomisto,

Poulsen, and Moran, 1998) in wet tropical sites. None of these studies have directly

addressed the role of disturbance.

The nature of disturbed habitats creates a positive feedback for species that prefer

disturbed sites. By their very nature such sites are unstable and result in increased

mortality due to continued habitat erosion. Pityrogramma ebenea is perhaps the most

common species of disturbed habitats at La Selva. The species produces large numbers of

spores with high fecundity, and the gametophytes can be found in virtually any habitat

where disturbance is present, i.e. exposed road/trail cuts and the relatively dark

understory (pers. obs.). The gametophytes of this species also germinate, grow, and

recruit rapidly (Fig. 2-6). Rapid development is necessary in species that occupy highly

disturbed habitats. Catastrophic events are common in the habitat of this species, and in

two populations such disturbance resulted in the near-complete habitat destruction and in

continued habitat instability exacerbated by wet season rains. Epiphytic species are also

subject to catastrophic disturbances; a single population of Vittaria lineata experienced

this sort of disturbance following a large tree fall which resulted in 100% mortality of one

of the study populations. However, these events seem relatively rare and epiphytic

habitats tend to be more stable when compared to terrestrial habitats in this forest.

Based on these data, terrestrial gametophytes simply do not establish in sites that

are disturbance-free. However, there are varying levels of tolerance and clearly different

life histories in terrestrial species. Danaea wendlandii is a eusporangiate fern. The

eusporangiates have many unique characters, but of particular interest for this study is the









production of liverwort-like gametophytes that are several cell layers thick. Individual

gametophytes are often large and more resistant to physical damage relative to the single

layered leptosporangiate species (pers. obs.). Such tough gametophytes that occur in sites

with minor disturbance confer longevity. Indeed, the gametophytes of Danaea

wendlandii exhibited 3 times the mean longevity and significantly less recruitment than

those of P. ebenea. Two populations ofDanaea wendlandii fell in the flood zone, and

while such disturbances are clearly part of the biology of this species, this event resulted

in lower mean longevities in this study. The eusporangiate biology of this species places

it nearly opposite of P. ebenea in terms of life history strategy.

There were also surprising differences between epiphytic and terrestrial species.

One emergent difference between these two groups is the percent of gametophytes that

survived but did not recruit. Over 70% of the gametophytes of C. brevifolium, and over

50% of both V lineata and L. vestita were alive and un-recruited by the 26th month. This,

compared to the less than 5% in Danaea wendlandii and 0% in Pityrogramma ebenea.

This observation highlights fundamentally different gametophytic life history strategies

that have evolved in the two life forms. In fact, recent phylogenetic analysis of the ferns

has revealed a recent split between terrestrial and epiphytic clades in the Eu-Polypodiales

One of the defining gametophytic characters of the epiphytic clade is indeterminate and

asexually reproducing gametophytes. Dassler and Farrar (1997) have argued that such

longevity and asexual reproduction is a mechanism to encourage outcrossing in epiphytic

species that may carry significant genetic load. Long-lived thalli and thus genotypes can

produce numerous archegonia over space and time to ensure fertilization of newly









dispersed genotypes. Such differences in life history are significant and may have been

critical in the radiation from terrestrial species into canopy habitats.

Density and Species Richness

Canopy gametophyte density is clearly more sensitive to disturbance, with nearly

58% occurring in undisturbed sites. Additionally, there was not a detectable relationship

between gametophyte density and light environment as was shown for terrestrial density.

In general, canopy light environments are significantly higher than terrestrial sites and the

lack of response to light in the former is not surprising in a habitat where this light is not

limiting. However, the canopy does experience temperature and humidity extremes and it

is plausible that microclimate and water availability play larger roles in these habitats

relative to terrestrial sites. Such an observation would be in line with reports by Hietz and

Briones (1998a) who demonstrated that within canopy distribution of fern sporophytes

was largely a function of species water relations.

Quantification of gametophyte species/morphotype richness in the natural

transects was abandoned largely due to time constraints. However, there was a clear

light-disturbance-density relationship for terrestrial species, and for this reason we

examined these variables more completely in the terrestrial disturbance experiment. In

one high disturbance plot, we counted six different morphotypes with 65% of the density

dominated by Pityrogramma ebenea. There are numerous life history strategies in the

ferns, and P. ebenea is a species with high germination and recruitment rates. Indeed, it is

possible that all of the species that were encountered in the higher disturbance plots

exhibited similar life histories. Clearly there are specific differences in gametophyte

ecology that influence densities at a given site. However, the near complete absence of

gametophytes in undisturbed habitats suggests that disturbance may be critical to the









majority, if not all terrestrial species regardless of life history. Such natural observations

combined with experimental manipulations offer strong evidence that light and

disturbance both act to structure terrestrial gametophyte density and richness.

Unlike comparisons of seedling-adult distributions, fern gametophytes are

completely and fundamentally different organisms from sporophytes. For this reason,

gametophytes would not necessarily be expected to behave like sporophytes. Firstly,

gametophytes are often, if not always more widespread than sporophytes (Peck, 1980;

Peck, Peck, and Farrar, 1990). Such plasticity and simplicity in function may reduce the

role that nutrients or other edaphic factors play in gametophyte distribution. Secondly,

gametophytes can be long lived with some individuals living decades in culture (Mottier,

1927; Walp, 1951) and in years field settings (pers. obs.). Additionally, gametophytes

may be relatively sensitive to desiccation and temperature changes; however, many

studies have shown that gametophytes are relatively robust to environmental stress (Sato

and Sakai, 1980; Cousens, 1981; Sato and Sakai, 1981; Cousens, Lacey, and Scheller,

1988; Ong and Ng, 1998). While survival from stress and edaphic requirements may be

important, for all species, emergence in litter free sites seems critical.

Sporophyte Ecology

Gametophyte and sporophyte distributions are related; however, the point of

gametophyte establishment is not necessarily the point of mature sporophyte distribution.

Epiphytic species are often associated with creeping rhizomatous growth. Such growth

may allow a perfectly healthy gametophyte to produce sporophytes in less than optimal

conditions. Such sporophytes may have the ability to then grow into more favorable

habitats where they can reproduce. The resultant mature sporophyte distributions

produced by such a strategy may obscure much of the species biology. Epiphytic species









may also form gametophyte banks that are long lived and stress tolerant and like many

seed banks, have the ability to wait for appropriate conditions to appear before recruiting

in to the sporophyte generation. This may be especially true for species whose

gametophytes also have a means of vegetative reproduction. This ability is common in

epiphytes (Atkinson and Stokey, 1964; Farrar, 1990; Farrar, 1998) and has been reported

in some terrestrial species (Watkins and Farrar, 2005). Such abilities question the

common assumption that gametophyte 'safe sites' limit the establishment of sporophytes

in epiphytic species. This hypothesis may hold more merit with terrestrial species but is

unlikely as important for epiphytic species.

Conclusions

Such apparent and fundamental differences in life history and the way that

epiphytic and terrestrial life forms respond to disturbance and light provides evidence for

adaptively meaningful variation in life histories that has evolved in the two groups.

Epiphytic species have evolved in a high light, highly competitive, yet relatively stable

matrix. Such environments reduce the light limitations encountered by terrestrial species,

yet they incorporate closer contact with bryophytes. Dassler and Farrar (1997) have

argued that differences in gametophyte morphology and asexual reproduction between

epiphytic and terrestrial species have largely evolved due to pressures form bryophyte

competition. Such changes may have only been possible in canopy habitats where

disturbance is less intense. Radiation into canopy habitats required a suite of adaptive

characters in both the gametophyte and sporophyte generation. One major advantage of

the canopy habitat is reduction in litter that can cover developing gametophyte. While

canopy habitats accumulate enormous amounts of organic matter (Cardelus and Chazdon,

2005), wind often removes a significant proportion of leaves that land on internal and









especially outer branches (Nadkarni and Matelson, 1991). The presence of leaf litter may

be the most important limiting factor to terrestrial species establishment as there are

relatively few morphological or physiological pathways that would allow a species to

survive under leaf litter. The mechanisms behind sporophyte distributions remain

complicated as they clearly also rely on gametophyte ecology. This is further complicated

by spore dispersal which may obscure the relationship between patterns of distribution

and habitat heterogeneity. Regardless of dispersal limitations or lack thereof, terrestrial

sporophytes are largely elements of disturbance past. Additional work on gametophyte

ecology will need to take into account edaphic and microclimatic factors to better

understand variables shaping the distribution of this magnificent group of plants.









Table 2-1. Relationship of gametophyte density and richness with three levels of
experimental disturbance and two light levels
Gametophyte density
Source DF F P>F

Light 1 52.643 0.000
Disturbance 3 18.567 0.000
Light*disturbance 3 6.152 0.001



Number of species
Source DF F P>F

Light 1 46.522 0.000
Disturbance 3 6.662 0.000
Light*disturbance 3 0.364 0.779














Table 2-2. Demographic and survival analyses for the gametophytes of 5 fern species using the Wilcoxon test to compare survival
distribution functions for different species


Number of Gametoohvtes


Survival Analysis


Mean
Species Dead Censored Total Months Std X2 P > X2
Campyloneurum brevifolium (Lodd. ex
Link) Link 19 87 106 22.9 0.57 386.2 <0.0001
Danaea wendlandii Rchb. f. 168 95 263 12.3 0.56
Lomariopsis vestita E. Fmyn. 22 45 67 20.8 0.82
Pityrogramma ebenea (L.) Proctor 165 110 275 4.1 0.28
Vittaria lineata (L.) Sm. 40 58 98 16.6 1.08












25


20
0)
r-
Z 15
(0
E
0
, 10-
-o
E
z


0 1 2 3
Level of Disturbance

25- B r2=0.65, F=53.58, p<0.001 a


20 -


15-


10-
b
b
5- b


n -


Level of Disturbance


Figure 2-1. Number of gametophytes counted and their relation to disturbance from
natural transects in both A) Canopy habitats and B) Terrestrial habitats. Here
0 represents no disturbance (< 5cm2 bare soil), 1 = low disturbance (i.e. >
5cm2 bare soil), 2 = medium disturbance (>5cm2 bare soil and soil disturbed),
3= high disturbance (100% bare soil and soil turned over).


A r2=0.11, F=2.40, p=0.08


a

JL


b b b



F~ ~
















60


50

U)
= 40
r-









10
n






0
S10




50

0)
0





CL60


50

0
40







20
10







10
0







0


IN BF ES PD UN AT

Type of Disturbance



B Terrestrial Habitats



















_


NL OT RT ER BF

Type of Disturbance


AT TU


Figure 2-2. The percentage of fern gametophytes as influenced by type of disturbance
identified. A) In canopy habitats (IN: insect damage, BF: branch fall, ES: epi-
slide, PD: physical damage, UN: Unknown, AT: animal trail) and B) In
terrestrial habitats (NL: no leaf litter, OT: old tip-up mound, RT: rotting
tree/wood, ER: erosion, BF: branch fall, AT: animal trail, TU: recent tip-up
mound)


A Canopy Habitats



















-- -- -


-


-


-


-


-


-



















A Canopy Habitats r2=0.000, p=0.085


1.5
*

1.0 --
** -
n *O ** *
*
0.5 o0 n e
*

0.0 m m m
I I .


1.4 1.5 1.6 1.7 1.8
Log (Canopy Openess)


2.0
C,,

S1.5 -
0
E
(D 1.0 -
4-
o
.3
E 0.5 -
z
0)
- 0.0-


1.9 2.0


B Terrestrial Habitats
**
0 EO
0 0



...
*emm *
*** e *
** m



e m r2=0.327, p=0.000


.0 0.5 1.0 1.5 2.0 2
Log (Canopy Openess)


Figure 2-3. The relationship between canopy openness and gametophyte density for A.
canopy and B. terrestrial species















1000


High Light
c'4 800 I I Low Light

V)

o. 600
o
E
(C
-400
0
-0
E
z 200



0 i 7-- I T
Control Low Med Hi

Level of Disturbance

Figure 2-4. Gametophyte densities as influenced by light and disturbance in experimental
plots. Control plots were those that had <5cm2 of bare soil exposed. Low
disturbance was created by only removing litter, medium plots were
established by removing litter and raking soil to a depth of 5cm, high plots
were created by removing litter and turning soil over with a spade to a depth
of 20cm















25



n 20


--
> 15
a

0
Z 10
10
(D


Campyloneurum Lomariopsis Vittaria Danaea Pityrogramma
brevifolium vestita lineata wendlandii ebenea


Campyloneurum Lomariopsis Vittaria Danaea
brevifolium vestita lineata wendlandii


Pityrogramma
ebenea


Figure 2-5. Mean longevity (months) A) and percent gametophytes still alive and un-
recruited B) for the 25-month period of the study















7/


\ .... ......

4 r ~- .'-.go. ---
0 .. -


\


Campyloneurum brevifolium

__, Lomariopsis vestita
S Vittaria lineata


1.0


S0.8-


B 0.6-
C)

5 0.4-
0
0

0.2-


0.0-


15 24 25 26
15 24 25 26


Time (Months)


S Pityrogramma ebenea
Lomariopsis vestita
Danaea wendlandii




Vittaria lineata
-- Campyloneurum brevifolium


L/ /
0 5 10 15 24 25 26

Time (Months)


Figure 2-6. Kaplan-Meier survivorship curves A) and proportion recruiting B) of 5 species
of fern gametophytes over the 25-month study period. No data were collected
during months 4-7 and 15-24


SDanaea wendlandii

---a Pityrogramma ebenea


0)
0 0.4

o

ry
. 0.2

0
o



0.0 *


I














CHAPTER 3
COMPARATIVE DESICCATION TOLERANCE OF TROPICAL FERN
GAMETOPHYTES: ECOLOGICAL AND EVOLUTIONARY CONSEQUENCES

Introduction

Overwhelming evidence indicates that land plants evolved from simple aquatic

algal ancestors (Bold, 1957; Niklas, 1997). The radiation of once-aquatic plants onto dry

land required the evolution of adaptive character suites that permitted life in what was

and remains deadly dry air. To survive this environment, plants have evolved two

mechanisms of surviving desiccating conditions. One mechanism is the avoidance of

desiccation as demonstrated in most modern terrestrial plants and has been accomplished

by the development of characters such as highly organized cuticles with effective

stomatal control. Cacti of the dry deserts perhaps represent the pinnacle of avoidance

with their succulent water storing stems, reduced leaves, and thick, well-developed

cuticle. These characters and many other associated with desiccation avoidance are all

thought to be highly derived and early land plants most certainly did not have such

structures. Early plants relied on the unique mechanism of survival from desiccation,

referred to as desiccation tolerance (DT). Desiccation tolerance has been differently

defined; the most commonly accepted definition is survival of drying to equilibrium with

surrounding air (Bewley, 1979). Such drying is more than sufficient to kill most any plant

that relies on desiccation avoidance: at least 99% of the world's vascular flora (Alpert

and Oliver, 2002).









Many of the characters that facilitated DT in ancestral land plants are still found

in modern algae and bryophytes (Alpert, 2000; Oliver, Tuba, and Mishler, 2000; Alpert,

2005). Much as the cacti are to avoidance, the bryophytes are to tolerance. Fantastic

stories exist in the literature demonstrating that some bryophytes can recover from 23

years of desiccation in herbaria (Alpert 2000 and references herein). Such remarkable

abilities clearly have ecological consequences and many studies have shown that more

desiccation tolerant bryophytes are often associated with more xeric habitats.

Consequences exist even within bryophyte species with one example being the

overrepresentation of female gametophytes in dioecious bryophytes of xeric habitats.

Such sex-based disparity is often related to greater DT in females relative to males (Stark

2005). Desiccation tolerance holds tremendous potential to influence the ecology of

species (Deltoro et al., 1998; Csintalan, Proctor, and Tuba, 1999; Robinson et al., 2000;

Cleavitt, 2002).

Apart from the linkage of DT to phylogeny, there have also been demonstrated

differences in desiccation tolerance of different generations, such as larval-adult in some

invertebrates or gametophyte-sporophyte in some bryophytes. The ability and degree of

desiccation tolerance in the different stages can be radically different. In some cases a

high degree of tolerance may exist in one stage but be absent in the other. For example,

the gametophytes of some bryophytes exhibit a much greater degree of tolerance than

sporophtyes (Proctor, 2000; Proctor, 2001). Such variation also occurs invertebrates

where the larvae of the fly Polypedilum vanderplanki exhibits greater desiccation

tolerance relative to the adult stage (Watanabe et al., 2002). Within the vascular plants,









desiccation tolerance of spores and seeds is well known but much less is known about

tolerance in lineages with two separate free-living stages.

The only lineage of vascular plants to exhibit two separate free-living generations

is the pteridophytes. Of particular interest to desiccation tolerance are the morphological

and physiological differences between these two stages in the ferns. The gametophyte is

the point of gamete formation and fertilization and is small, lacks vascular tissue, and has

a poorly developed to non-existent cuticle. The sporophyte produces spores and is thus

the primary stage for dispersal; it has a well developed vascular system and a waxy

cuticle complete with stomata. These differences alone result in unique life-cycle-

mediated ecological strategies, especially as they relate to water relations. True

desiccation tolerance in the sporophyte stage is known from and likely only exists in

relatively few species (Gaff, 1987; Porembski and Barthlott, 2000). In a recent review on

the subject, Proctor and Pence (2002) recorded that <1% (64 species) of the ferns studied

exhibited DT and of those, 40 were Cheilanthoid taxa that are commonly associated with

desert-like habitats. Much less is known of species from tropical habitats, but DT has

been recorded in genera as phylogenetically disparate as Asplenium and Polypodium

(Kappen, 1964; Gaff, 1987; Proctor and Pence, 2002). The species in which it does occur

are often extreme xerophytes living in deserts or other highly exposed and dry

environments. As with bryophytes, the apparent degree of desiccation tolerance in the

sporophyte stage has been linked with species ecological distribution (Harten and

Eickmeier, 1987; Hietz and Briones, 1998b). Studies are still too sparse to determine the

extent of this character in structuring populations.









Studies on desiccation tolerance of the gametophyte generation of the ferns are

fewer in number. Some of the earliest comments were made by Goebel (1900) regarding

the ability of the buried tubercles of Annogramme chaerophylla to resume growth

following dry spells. A similar observation was made by Cambell (1904) on the unburied

gametophytes of Gymnogramme triangularis that appeared to have survived a dry

California summer. The first evidence was generated by Pickett (1913; 1914; 1931) who,

through a series of desiccation experiments was the first to clearly show that the

gametophytes ofAsplenium rhizophyllum and A. platyneuron could recover growth

following extreme desiccation. He also discovered that there was a greater degree of

tolerance in A. Rhizophyllum, a species of more exposed and drier habitats, relative to A.

platyneuron, which is often confined to more mesic sites. This was the first link of

desiccation tolerance in the gametophyte generation with sporophyte distributions and

species ecology. Pickett's work has largely been the last of its kind, and apart from

anecdotal reports (Gilbert, 1970) and observations on the ability of the gametophytes of

Pyrossiapilosellodes to recover from drought (Ong and Ng, 1998) nothing is known of

the ability of fern gametophytes to tolerate desiccation and of their rates of recovery.

The goal of this paper is to survey a broad range of tropical fern gametophytes to

determine the extent of desiccation tolerance in this phase of the fern life cycle. We first

examine the ability of several species to recover from a single desiccation event. We then

subject a select number of several species of varying life histories to a more extensive

repeated dry down cycles and drying intensities and relate the results to the ecological

distributions of the species.









Materials and Methods

Spore Material and Growth Conditions

Spore material from 12 species of varying ecology was collected from La Selva

Biological Station in the Atlantic lowlands of northeastern Costa Rica at 37-100 masl.

Fertile fronds were gathered in the field and put into glassine envelopes with tape-sealed

seams. Envelopes with plant material inside were stored in an air-conditioned lab and

allowed to dry under these conditions. Spores were brought back to the University of

Florida where they were cultured. The growth of a broad sampling of species with

different ecologies required extensive experimentation with culture techniques, it was

discovered that species grew best on a combination of organic soil collected from canopy

trees at La Selva mixed with a small amount of vermiculite. Spores were sown on this

medium into 60mm x 15mm Fisherbrand Petri plates. These plates were stored in sealed

clear plastic containers (Pioneer Plastics Model 395-c, Dixon, KY). Cultures were

exposed to 20tmol m-2 sec-1 for 10hrs day- from GE fluorescent plant and aquarium

40watt growbulbs and watered with deionized water every 10-12 days.

Desiccation Experiments

For the initial survey experiment, 5-10 mature gametophytes (one gametophyte

per Petri plate) of all 12 species (see Table 3-1) were allowed to desiccate at a vapor

pressure deficit (VPD) =1.3kPa (50% relative humidity) in a VPD controlled chamber

that was constructed using one of the plastic growth boxes connected via Bevline tubing

to a Licor dew point generator (Model 610, Lincoln, NE) set to a flow rate of 0.5 L min'.

Samples were allowed to dry for 45min in a constant vapor pressure deficit and were

removed from the box every 5min and placed in a Sartorious microbalance (Gottingen,

Germany) where weight and a measurement of Fv/Fm was taken (see methods below). The









samples were then placed back into the box. The volume of the box was relatively small,

and a Hobo Pro RH/Temp Data Logger (Bourne, MA) was used to verify that the

container typically regained the 1.3kPa VPD within one min of the top being replaced. To

evaluate recovery, upon completion of the desiccation treatment, samples were

consecutively rehydrated by adding 5-10 drops of deionized water to the thallus. After 1

hr of rehydration, measurements of Fv/Fm were again made at 5min, 24hrs, and 48hrs post

rehydration. Samples were then dried for 72hrs in a drying oven at 700C and weighed to

determine gametophyte dry weight. Relative water content was plotted against time and

Fv/Fm.

A second desiccation experiment was designed to test the effect of drying

intensity on recovery of Fv/Fm. The gametophytes ofDiplazium subsilvaticum and

Phlebodium pseudoaureum, and Polypodium triserale were chosen to represent the

extremes of tolerance from the initial desiccation experiment and they were dried at three

different intensities VPD=0.5kPa (20% RH), VPD=1.3kPa (50% RH) and VPD=2.lkPa

(80%RH) following the methods above. Gametophytes were kept at these VPD for 72hrs

after which time they were rehydrated with deionized water and measurements of Fv/Fm

were taken at 24, 48, and 72hr post rehydration. These values were related to the dark

adapted value of Fv/Fm to determine the mean percent recovery.

A third experiment was run to examine the influence of consecutive desiccation

cycles on photochemical efficiency. Plant material from six species was chosen from the

survey experiment to represent the different recovery abilities. Thirty gametophytes were

selected and ten were dehydrated for one, two, or three cycles at VPD=1.3kPa and kept at

this level for 72 hrs using the methods described above. Material was then rehydrated









with deionized water and measurements of Fv/Fm were again made at 24hrs, 48hrs, and

72hrs post rehydration. These values were related to the dark adapted value of Fv/Fm to

determine the mean percent recovery.

Chlorophyll-Fluorescence Measurements

Variation in photochemical efficiency (Fv/Fm) was measured as the desiccation

dependent change in ratio of variable and maximal fluorescence Fv/Fm where Fv is the

difference between the maximum (Fm) and the minimum (Fo) fluorescence emissions

(Mulkey and Pearcy, 1992; Horton, Ruban, and Walters, 1996) measured using an Opti-

Sciences pulse modulated fluorometer (Model OS-500, Hudson NH). Minimal

fluorescence was measured under a weak pulse of modulating light over 0.8 s, and

maximal fluorescence was induced by a saturating pulse of light (8000 imol m-2 -1)

applied over 0.8 s. The parameter Fv/Fm was first measured after 20mins dark adaptation,

and this measurement was taken as the index of recovery. Dark-adapted FV/FM provides

an estimate of the maximal quantum efficiency of Photosystem II, which in unstressed

material is generally around 0.76-0.83 (Demmig-Adams and Adams, 1992).

Statistical Analysis

For the initial desiccation survey, a series of regressions was run on arcsin square root

transformed relative water content (RWC) data against time for each individual within a

species to determine the rate of drying of gametophytes exposed to a VPD of 1.3kPa

(50% RH) over the 45min time interval. The slopes of these regression lines were

calculated to generate a species mean drying rate. These rates were then analyzed by a

one way ANOVA followed by a post hoc Tukey's test to determine differences among

species. Linear regression analysis was used to asses the influence both final relative and

absolute water content at 45min and the slope of the individual drying curves on species









recovery ability at 48h post rehydration. Percent species recovery data were also arcsin

square root transformed for these analyses. The mean species drying rates, RWC, and

absolute water content (AWC) at 45min were then plotted against each species' mean

percent recovery at 48h. Depression in photochemical efficiency was also graphed as a

function of thallus RWC and AWC.

To assess the influence of consecutive desiccation cycles (experiment 3) and VPD

(experiment 2) on photochemical efficiency, a repeated measures ANOVA was

performed with number of desiccation cycles or VPD and recovery time as the fixed main

effects. Data were first examined for sphericity following the Mauchly criterion. Pairwise

comparisons were made across recovery times with Bonferroni-adjusted multiple t-tests.

Klockars and Sax (Klockars and Sax, 1986 p. 38-39) recommend using the more

stringent Bonferroni-adjusted multiple t-test when the number of planned comparisons is

greater than the number of degrees of freedom for between-groups. In cases where data

did not meet the sphericity criterion, p-values were adjusted using both Greenhouse-

Geisser and Huynh-Feldt methods based on the respective epsilons (Scheiner and

Gurevitch, 2001).

Results

Desiccation Survey

For the initial desiccation survey, a series of regressions were run on each species

to determine the change in relative water content (RWC) of gametophytes exposed to

1.32kPa (50% RH) over the 45min time interval. All species exhibited rapid rates of

thallus relative water content loss (Fig. 3-1A) and absolute water content loss (data not

shown). In all species regressions on the arcsin square root transformed data were linear

(Table 3-1). In all cases, linear regressions were used to calculate the slopes of species'









RWC and AWC drying curves to determine desiccation rates. The absolute size (mass

based) of individual gametophytes had little influence on the absolute rate of water loss

(Fig 3-3a, r2=0.05, p=0.06). These rates varied significantly among species with the

fastest dry down in the terrestrial Thelypteris balbisii and slowest rates in the terrestrial

Cyclopeltis semicordata and the epiphyte Polypodium triserale (Fig. 3-3A). Depression

in photochemical efficiency (Fv/Fm) as gametophytes desiccated was non-linear with

respect to RWC and varied among species (Fig. 3-1B).

Species exhibited differential abilities to recover following desiccation (Fig 3-

3B). This recovery ability was more closely related to the decay rate of RWC (r2=0.288,

p<0.0001) when compared to the final RWC reached (r2=0.193, p=0.0008), the decay

rate of AWC (r2=0.0001, p=0.81), or final AWC reached (r2=0.097, p=0.008) after

45min.

Desiccation Rates

The VPD of the different desiccation treatments significantly influenced the

recovery abilities of both Diplazium subsilvaticum and Phlebodium pseudoaureum, but

had little influence on Polypodium triseriale. For the understory terrestrial D.

subsilvaticum, the ability to recover following the 2.12 and 1.3kPa VPD treatments was

essentially non-existent. Additionally, the Fv/Fm values reached at these VPDs are

suggestive of significant photoinhibition and photodamage. The 0.53kPa treatment also

depressed Fv/Fm but to a lesser degree and gametophytes exposed to this treatment

exhibited clear recovery following rehydration. Phlebodium pseudoaureum exhibited

relatively less depression in Fv/Fm and exhibited greater recovery than Diplazium

subsilvaticum. In all three VPD treatments, gametophytes exhibited recovery albeit with

lower rates from the 2.12 and 1.32kPa treatments. Polypodium triserale exhibited









remarkable Fv/Fm fidelity at all three rates with no significant degree of Fv/Fm

depression at any desiccation intensity (Fig. 3-5, Table 3-2).

Desiccation Cycles

Six species representing different life histories and desiccation tolerance from the

initial survey were chosen and exposed to multiple desiccation cycles (1, 2, or 3). In all

cases excluding Polypodium triserale, percent recovery was greater following one versus

two or three desiccation cycles (Fig. 3-6 f, Table 3-3). Recovery ability was closely

linked to species ecology with slow to no recovery following >1 desiccation cycle in the

understory species that occur in more mesic habitats: Diplazium subsilvaticum, Adiantum

latifolium, and Cyclopeltis semicordata. Within this group, Cyclopeltis semicordata

exhibited less inhibition and the greatest recovery following a single desiccation cycle.

With multiple cycles, all species were significantly inhibited and there was little evidence

of recovery following 2 and 3 cycles. There was a much greater degree of recovery

following 2 and 3 cycles in the species from more xeric habitats: the terrestrial

Pityrogramma ebenea, and the epiphytes: Phlebodium pseudoaureum, Polypodium

triserale. Unlike the other species in this xeric category, Polypodium triserale exhibited

similar degrees of recovery from 1 and 2 cycles but experienced a much slower recovery

following the third desiccation cycle.

Discussion

In all species, fern gametophytes exhibited rapid rates of water loss (Fig la.).

With poor control of transpiration and inefficient absorptive organs, gametophytes likely

rely on water derived directly from the atmosphere and/or that which flows over them. As

such, gametophytes face considerable variations in water content throughout the day and

must be able to withstand long periods of desiccation. This is especially true of epiphytic









gametophytes that can live for years (Watkins et al. Chapter 2). The only recourse that

gametophytes have is to tolerate desiccation or perish.

The initial survey produced a surprising degree of desiccation tolerance across

many species in a gametophyte that is known to require water for fertilization and is

thought to require humid conditions for survival. After exposure to a VPD of 1.3kPa

(rH=50%) for 24 hours, all species exhibited greater than 50% recovery of the pre-

treatment Fv/Fm values and the majority had recovered more than 70% of this value (Fig.

3-3). Gametophytes were dried to near constant state and thus match the definitions of

desiccation tolerant by Bewley (1979) and Alpert and Oliver (2002). The species in this

study are all tropical in origin and while they experience various degrees of humidity in

nature, some of the more exposed canopy trees for this same forest average VPD-3.0kPa

at mid-day during the dry season (Cardelus and Chazdon, 2005). The gametophytes in

this study experience VPD levels of 1.3kPa, but the value remains on the extreme end of

what they typically experience. There was no significant difference in the absolute water

loss rates based on gametophytes mass (Fig. 3-2). In natural settings, individuals will be

exposed to daily variation in RWC and their ability to recover from low relative water

content is crucial. The only apparent mechanism that gametophytes have to control water

loss is an increase in size and perhaps alteration of thallus morphology (see below).

Variation in VPD

To further examine the influence of desiccation intensity on recovery of

pretreatment photochemical efficiency, three species were exposed to three different

desiccation intensities: roughly the everyday vapor pressure deficient in nature (0.53kPa),

that which is likely to reflect a typical drought event (1.32kPa) and an extreme value

representing a VPD that species in this site rarely if ever experience (2.12kPa). The









results from this experiment demonstrated remarkable tolerance to desiccation intensity

that is tightly linked to species ecology. Diplazium subsilvaticum is a low-light creek-side

species and has little tolerance of desiccation intensities below 0.53kPa (80% RH) (Fig 3-

5A). The biggest decrease in Fv/Fm occurred in this species between 0.53 and 1.32kPa.

The two epiphytes occur in similar habitats; and, Phebodium pseudoaureum is typical of

open and exposed habitats such as roadsides and open clearings, Polypodium triseriale is

often found in more highly exposed areas such as fence posts and tree trunks. The level

of tolerance to desiccation intensity was linked to the habitats with the most highly

exposed Polypodium triseriale exhibiting essentially no sensitivity to desiccation

intensity (Fig. 3-5C).

These patterns enforce the notion that desiccation rates can influence the survival

of certain species. In some bryophytes (Gaff, 1997; Alpert and Oliver, 2002; Alpert,

2005) and at least in the sporophtyes of the Hymenophyllaceae (Proctor, 2003),

intermediate desiccation intensities that maintain intermediate RWC's are more likely to

result in mortality compared to rapid dry downs. Alpert and Oliver (2002) have argued

that one reason for this pattern is need for rapid yet organized metabolic shutdown that

may not occur at slower desiccation rates. Additional studies that incorporate different

drying intensities and longer time spent in these conditions are clearly needed.

The link between DT and species distributions corresponds closely with those

reported for bryophyte species from xeric habitats exhibiting greater desiccation tolerance

than those from mesic habitats (Oliver, Mishler, and Quisenberry, 1993; Deltoro et al.,

1998; Proctor, 2001; Cleavitt, 2002; Alpert, 2005). It also suggests that there may be

some connection between gametophyte and sporophyte physiologies.









Desiccation Cycles

Not only do species experience different intensities of desiccation in nature, they

also experience multiple desiccation cycles throughout the day and/or growing season.

The species in this study clearly exhibited different abilities to cope with consecutive

desiccation cycles at a VPD of 1.3kPa with species of more mesic habitat exhibiting little

ability to cope with more than one cycle of desiccation (Fig. 3-6). The more mesic

creekside Diplazium subsilvaticum was the most desiccation-sensitive after one cycle and

had the worst recovery whereas the more xeric Adiantum latifolium and Cyclopeltis

semicordata had higher recoveries following one cycle (Fig. 3-6 c&b). Polypodium

triseriale, the species of more xeric habitats also exhibited depression in Fv/Fm, but

recovery was largely independent of two desiccation cycles. One aspect that was

common for all terrestrial species is the relative tolerance to a single desiccation cycle

and the extreme depression caused by repeated cycles. Repeated desiccation cycles

likely induce radical biological damage and recovery depends more on actual DNA repair

and new protein synthesis than simple recovery of PSII function related to the release of

excess excitation energy. Light and desiccation combined are often a deadly combination;

and that gametophytes were returned to the original culture conditions upon rehydration

could have resulted in increased photodamage. This was however similar to what species

experience in nature and is likely reflective of species biology.

The ability of species to recover from desiccation was more closely related to the

rate of drying rather than the final RWC (Fig 3-4). While gametophytes initially seem to

have relatively few options to control the rate of thallus desiccation there was variation in

rates among species exposed to identical drying conditions (Fig 3-3a). For example, the

terrestrial Thelypteris nicaraguensis a RWC decay rate that was more than twice as fast









as the canopy epiphyte Polypodium triseriale. The variation in desiccation rate was

linked to gametophyte morphology (Fig. 3-4). Species with complex three-dimensional

morphologies or those that tended to produce prothallia hairs exhibited significantly

slower dry down rates than gametophytes that are one dimensional and glabrous. This

observation suggests a novel mechanism that gametophytes may employ to control water

loss. There is limited internal capacitance in fern gametophytes, and in a manner similar

to many bryophytes, fern gametophytes with even a minor degree of morphological

complexity can hold external water. Such exohydric abilities may function in a natural

setting to help slow the rate of water loss. The decrease in rate was more likely a

combination of water vapor becoming trapped in the folds and overlapping wings of

more complex thalli and the increase of external boundary layer produced by

gametophyte proliferations and hairs.

Variation in gametophyte morphology especially in complexity has long been

commented upon. Dassler and Farrar (1997, 2001) have speculated that complex

morphologies found in many long-lived gametophytes of epiphytic species have arisen

from competition with bryophytes and as a mechanism to ensure outcrossing. One

obvious trend across the taxa is that gametophytes of species from drought-prone habitats

such as those in epiphytic habitats, deserts, rock outcroppings, etc., tend to produce thalli

that often exhibit complex branching, overlapping wings, proliferation, and hairs. Apart

from this link between gametophyte morphology and ecology there is also a link between

gametophyte morphology and phylogeny. Recent phylogenetic analysis of the ferns has

revealed a split between terrestrial and epiphytic clades in the Eu-Polypodiales (Pryer,

Smith, and Skog, 1995). The athyrioids, thelypteroids, onocleoids, woodsioids and









blechnoids are almost entirely terrestrial; perhaps less than 1% of the known species of

this clade are true epiphytes. On the other hand, the dryopteroids, lomariopsoids,

elaphoglossoids, oleandroids, davallioids, and polypodioids have many epiphytic species;

perhaps as many as 60-70% of the species in this group as a whole are epiphytic (Fig 3-6)

(R. Moran pers. comm.). Most epiphytic species exhibit complex morphologies whereas

terrestrial species often less complex ones. Increases in thallus size and complex three

dimensional morphologies may provide the only water conservation mechanisms

available to fern gametophytes. Complex water conserving morphology may have been

critical in the radiation of ferns into canopy and more exposed habitats.

Conclusions

The data presented in this paper show remarkable desiccation tolerance in fern

gametophytes. While all species exhibited recovery following an extreme desiccation

event, the extent of recovery differed among species and was closely linked to the

ecology of the species. The role of the fern gametophyte in controlling recruitment

remains unclear, but these data suggest that gametophytes are relatively robust in dealing

with desiccation especially in limited cycles. While desiccation intensity clearly

influenced recovery, repeated cycles of desiccation were more likely to limit recovery

and in the case of the most mesic species likely resulted in significant photodamage. The

degree of recovery following desiccation and its relation to species ecology suggest that

fern gametophytes exhibit adaptively meaningful variation in this character. It is likely

that selective pressures acting on the gametophyte are largely responsible for the

distribution of ferns and have played a major role in the evolution of ecological diversity

within the group.














Table 3-1. Species and life form from the initial desiccation survey, a series of regressions were run on each species to determine the
change in relative water content (slope) of gametophytes exposed to 1.32kPa (50% rH) over the 45min time interval. All
species exhibited rapid and uncontrolled rates of thallus water loss. In all species regressions on the arcsin square root
transformed data were linear (STE are standard errors, RWC is relative water content expressed as ((g fresh weight- g dry
weight)/g saturated weight g dry weight))* 100, AWC is absolute water content expressed as mg water / mg dry mass, %
rec corresponds to percent recovery of initial pre-treatment dark adapted Fv/Fm values


Species

Adiantum latifolium Lam.

Cyclopeltis semicordata (Sw.) J. Sm.
Dennstaedtia bipinnata (Cav.)
Maxon

Diplazium subsilvaticum H. Christ

Nephrolepis biserrata (Sw.) Schott

Phlebodium pseudoaureum (Cav.)
Lellinger

Pityrogramma ebenea (L.) Proctor

Polypodium triseriale Sw.

Pteris altissima Poir.
Thelypteris balbisii (Spreng.) Ching

Thelypteris curta (H. Christ) C. F.
Reed

Thelypteris nicaraguensis (E. Fourn.)
C. V. Morton


Life Form r2 p Slope STE RWC STE AWC STE %rec STE


Terrestrial

Terrestrial

Terrestrial

Terrestrial

Terrestrial


Epiphyte

Terrestrial

Epiphyte

Terrestrial
Terrestrial


0.755

0.811

0.939

0.862

0.882


0.913

0.878

0.973

0.834
0.816


<0.0001

<0.0001

<0.0001

<0.0001

<0.0001


<0.0001

<0.0001

<0.0001

<0.0001
<0.0001


Terrestrial 0.711 <0.0001


Terrestrial 0.762 <0.0001


-1.891

-1.359

-1.670

-1.512

-1.419


-2.033

-2.017

-1.279

-1.611
-1.490


0.101

0.074

0.045

0.084

0.094


0.059

0.081

0.203

0.090
0.120


7.375

34.255

19.858

31.880

34.858


9.646

13.737

49.021

23.856
32.866


2.138

3.836

2.777

5.680

4.298


1.449

3.048

8.097

9.015
7.795


1.340

5.660

2.320

3.640

5.540


2.530

5.370

8.640

9.490
3.130


0.164

0.820

0.290

0.980

1.170


0.500

0.670

1.471

0.981
0.211


0.579

0.897


0.739

0.705

0.823


0.678

0.499

0.807

0.771
0.828


0.116

0.019


0.072

0.041

0.056


0.055

0.100

0.065

0.065
0.070


-1.463 0.116 18.875 5.829 3.380 0.794 0.600 0.127


-2.199 0.061 19.254 3.512 3.690 0.370 0.555 0.087












Table 3-2. Fv/Fm recovery results from the repeated measures ANOVA for gametophytes exposed to three different desiccation
intensities: 20%RH (-0.53kPa), 50%RH (-1.32kPa) and 80%RH (-2.12kPa). The gametophytes of Diplazium
subsilvaticum are often found in the understory, whereas those of Phlebodium pseudoaureum, and Polypodium triserale
were collected in the mid and exposed canopy respectively. Gametophytes were kept at the VPD levels for 48hrs after
which time they were rehydrated with deionized water and measurements of Fv/Fm were taken at 24, 48, and 72hr post
rehydration. These values were related to the dark adapted value of Fv/Fm to determine the mean percent recovery


Diplazium striatastrum Lellinger df F p Mauchly x p

VDP 2 38.41 <0.0001 0.739 3.24 0.662
ID(VPD) 12 1.86 0.0741
Recovery Time 2 7.32 0.0033
Recovery Time*VPD 4 6.39 0.0012

Phlebodium pseudoaureum (Cav.) Lellinger df F p Mauchly x p

VDP 2 42.19 <0.0001 0.716 3.59 0.61
ID(VPD) 12 2.2 0.0482
Recovery Time 2 24.31 <0.0001
Recovery Time*VPD 4 7.81 0.0003

Polypodium triseriale Sw. df F p Mauchly x p

VDP 2 1.46 0.2716 0.779 2.67 0.75
ID(VPD) 12 4.53 0.3531
Recovery Time 2 0.6 0.6352
Recovery Time*VPD 4 5.09 0.0041












Table 3-3. Fv/Fm recovery results from the repeated measures ANOVA for gametophytes exposed to 1, 2, or 3 desiccation cycles at
50% RH (VPD- 1.32kPa). Gametophytes were kept at this level for 48 hrs. Material was then rehydrated with deionized
water and measurements of Fv/Fm were again made at 24hrs, 48hrs, and 72hrs post rehydration. These values were related
to the dark adapted value of Fv/Fm to determine the mean percent recovery. Adjusted p-vaules are G-G Greenhouse-Geisser
and H-F Huynh-Feldt adjusted probabilities
Adjusted p
Diplazium striatastrum Adjusted
Lellinger df F p Mauchly X p G-G H-F
Rate 2 3.07 0.0649 0.6301 5.081 0.0788 0.0043 0.0016
ID(Trt) 12 3.58 0.3931
Recovery Time 2 98.7 <0.0001 GG E = 0.73
Recovery Time*Trt 4 4.06 0.0118 HF E = 0.944

Adiantum latifolium Lam. df F p Mauchly X p G-G H-F
Rate 2 1.53 0.2378 0.691 4.06 0.1312 0.0073 0.0029
ID(Trt) 12 2.09 0.4982
Recovery Time 2 324.53 <0.0001 GG E = 0.764
Recovery Time*Trt 4 5.44 0.0029 HF E = 0.999

Cyclopeltis semicordata (Sw.)
J. Sm. df F p Mauchly X p G-G H-F


Rate
ID(Trt)
Recovery Time
Recovery Time*Trt


600.01
1.98
7.33
2.43


<0.0001
0.5498
0.0083
0.753


0.706


3.827


0.1475 0.018


GGE
HF E


0.773
1


0.0094












Table 3-3. Continued
Pityrogramma ebenea (L.)
Proctor df F p Mauchly 2 p G-G H-F
Rate 2 57.65 <0.0001 0.857 1.688 0.4298 <0.0001 <0.0001
ID(Trt) 12 0.56 0.792
Recovery Time 2 80.09 <0.0001 GG E = 0.875
Recovery Time*Trt 4 11.67 <0.0001 HF E = 1

Phlebodium pseudoaureum
(Cav.) Lellinger df F p Mauchly 2 p G-G H-F
Rate 2 98.81 <0.0001 0.462 8.48 0.0143 <0.0001 <0.0001
ID(Trt) 12 0.38 0.8699
Recovery Time 2 169.52 <0.0001 GG E = 0.65
Recovery Time*Trt 4 20.49 <0.0001 HF E = 0.818

Polypodium triseriale Sw. df F p Mauchly 2 p G-G H-F
Rate 2 24.82 <0.0001 0.492 7.78 0.02 0.0046 0.0019
ID(Trt) 12 1.65 0.548
Recovery Time 2 87.4 <0.0001 GG E = 0.664
Recovery Time*Trt 4 6.78 0.0008 HF E = 0.839































0 10 20 30 40 5C
Time (min)

B
I|


100 80 60 40 20 0
Relative Water Content

-o- Thelypteris nicaraguensis
-o- Thelypteris curta
-- Pityrogramma ebenea
-- Adiantum latifolium
-A- Dennstaedtia bipinnata
-- Pteris altissima
--- Diplazium subsilvaticum
-*- Thelypteris balbisii
-*- Phlebodium pseudoaureum
-*- Nephrolepis biserrata
-- Cyclopeltic semicordata
-*- Polypodium triseriale

Figure 3-1. (a) Gametophyte drying curves from 12 tropical fern species of different
habitats. Species were exposed to a VPD to 1.32KPa (-50%RH) for 45 min.
(b) Depression of photochemical efficiency in the same gametophytes over a
series of decreasing thallus water contents





















0.0





-* -0.32 % *
ao e* e


QS *


C.)
-0.4 a


3 -0.5

-0.6


-0.7 -


-0.8
0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035 0.0040

Dry Mass (mg)


Figure 3-2. The rate of absolute water loss relative to gametophyte size as indexed by dry
mass















-3.0


-2.5 -


-2.0 -


-1.5 -


-1.0 -


-0.5-


0.0




1.0



0.8


0.6 -



0.4-


B
. x ^ ..


0 .0 i i I i I I Ii Ii I


Species
Figure 3-3. (a) Rate of thallus drying as calculated from Figure 3-1 for 12 tropical fern
species of different habitats. (b) Proportional recovery of the pre-treatment
dark adapted value of Fv/Fm in these same species


A F=10.23, p<0.0001


f
e e
_T_

t b b b b

n81_


















1.0
E
LL
0.9
LL


E 0.8


0
I-






0
ad 0.7



0
0.6
o
, 0.5

0.
r0
- 0.4

LL
0.3
-0.5


-2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0


Change in RWC min-




Figure 3-4. Proportional recovery of the pre-treatment dark adapted value of Fv/Fm and
rate of thallus water loss expressed as (a) relative water content ((g fresh
weight- g dry weight)/g saturated weight g dry weight))* 100 and (b)
absolute water content (g wet weight/g dry weight)


A






PteAlt








PitEbe


CycSem
F I


-0.4 -0.3 -0.2 -0.1

Change in AWC min1


0.3 -
-2.4


CycSem


-


5
















0.8 A


0.6 a a a

ILL
LL 0.4 b
b b
0.2 b b

Diplazlum subsilvaticum (understory)
0.0
Dark 24h 48h 72h

0.8 -B

b b b
0.6 b


L 0.4


0.2
Phlebodium pseudoaureum (Mid-Canopy)
0.0
Dark 24h 48h 72h


0.8 C


0.6
E
ILL
2> 0.4


0.2

0.0


Dark 24h 48h 72h
Time (h)



Figure 3-5. Fv/Fm recovery graphs for gametophytes of (a) Diplazium subsilvaticum, (b)
Phlebodium pseudoaureum, (c) Polypodium triserale exposed to three
different desiccation intensities: VPD-0.53kPa (20%RH), VPD-1.32kPa
(50%RH), and VPD -2.12kPa (80%RH). The gametophytes of Diplazium
subsilvaticum are often found in the understory, whereas those of Phlebodium
pseudoaureum, and Polypodium triserale were collected in the mid and
exposed canopy respectively. Gametophytes were kept at the VPD levels for
48hrs after which time they were rehydrated with deionized water and
measurements of Fv/Fm were taken at 24, 48, and 72hr post rehydration.
Pairwise comparisons were made across recovery times with Bonferroni-
adjusted multiple t-tests


-- 2 12 Kpa(20% Rh)
-0- 1 32 Kpa (50%r Rh)
-V- 0 53 Kpa (80% Rh)
Polypodium trlserale (Exposed Canopy)







53




100oo A Diplazium subsilvaticum loo D Pityrogramma ebenea

80 Cycle 1 80 a
-0- Cycle 2
8 -0- Cycle 3
S60 60
a
S40 a a 40
)c b
20 b b b 20-

0 Vb b 0
24 48 72 24 48 72

100 B Adiantum latifolium 100 E Phlebodium pseudoaureum
a
80 80 -
8 a b
I 60 a --- 60

8 40 40 b
a a
20 20
b bb
0 b__ ____________ b 20
24 48 72 24 48 72
100
C Cyclopeltis semicordata 100 F Polypodium triseriale a
80 a a a a
8 60 -

S60 -

40 b b 40
b 4 0
20 c 20

0- 0
24 48 72 24 48 72
Recovery Time (hrs) Recovery Time (hrs)


Figure 3-6. Proportional Fv/Fm recovery results for gametophytes exposed to 1, 2, or 3
desiccation cycles at VPD- 1.32kPa (50%RH). Gametophytes were kept at
this level for 48 hrs. Material was then rehydrated with deionized water and
measurements of Fv/Fm were again made at 24hrs, 48hrs, and 72hrs post
rehydration. These values were related to the dark adapted value of Fv/Fm to
determine the mean percent recovery. Pairwise comparisons were made within
recovery times with Bonferroni-adjusted multiple t-tests. (a) Diplazium
subsilvaticum, (b) Adiantum latifolium, (c) Cyclopeltis semicordata, (d)
Pityrogramma ebenea, (e) Phlebodium pseudoaureum, (f) Polypodium
triseriale (see Table 3-1 for species habitat information)











Polypodiales

i iHe AHemidictyoid
Cystopteridoid
""'4 (helypteridoid
Woodsioid
/ Blechnoid
Onocleoid
Athyrioid


Hypodematium
Didymochlaena -
Rumohrioid
Dryopteridoid
Cyclopeltis
Lomariopsis
$ : Nephrolepis
E
Tectarloid
Oleandra
Davallioid
Polypodioid


Figure 3-7. Morphology in fern gametophytes is diverse and is closely related to species
ecology and phylogeny. Gametophytes of species from drought prone habitats
such as those in epiphytic habitats, deserts, rock outcroppings etc. tend to
produce thalli that often exhibit complex branching, overlapping wings,
proliferation and hairs; whereas, species from more buffered habitats have less
ornamented and simple morphologies. The athyrioids, thelypteroids,
onocleoids, woodsioids and blechnoids are almost entirely terrestrial, perhaps
less than 1% of the known species are true epiphytes. On the other hand, the
dryopteroids, lomariopsoids, elaphoglossoids, oleandroid, davallioid and
polypodioids have many epiphytic species, perhaps as many as 60-70% of the
species in this group as a whole are epiphytic (R. Moran pers. comm.). Most
epiphytic species exhibit complex morphologies whereas terrestrial species
often less complex ones. Such differences in morphology are significant and
may have been critical in the radiation from terrestrial species into canopy
habitats. A) Adiantum latifolium, B) Thelypteris sp. 1, C) Thelypteris sp.2, D)
Vittaria, E) Campyloneurum














CHAPTER 4
NITROGEN-15 NATURAL ABUNDANCE AND NITROGEN USE STRATEGIES OF
THE GAMETOPHYTES AND SPOROPHYTES OF TROPICAL EPIPHYTIC AND
TERRESTRIAL FERNS

Introduction

A central goal of plant ecology is to develop a mechanistic understanding of

species distributions in both space and time. One important abiotic factor that is known to

influence both plant performance and distribution is nitrogen. In N-limited systems or in

systems where N availability is heterogeneous, plants can compete for N in many ways,

one of which is by partitioning this resource in both space and time or by uptake of

different chemical forms of N. Such partitioning may result in species coexistence

through sorting along resource gradients which ultimately structures species distributions.

The partitioning and uptake of different chemical forms of N has been receiving

increased attention largely due to the revelation that plants can circumvent the N cycle

and directly uptake organic N, in the form of amino acids from the soil solution (Chapin,

Moilanen, and Kielland, 1993a; Kielland, 1994, 1997; Lipson and Nasholm, 2001; Finzi

and Berthrong, 2005). While it has been known for some time that plants can acquire

organic N, early work in tundra and boreal ecosystems demonstrated that a large

component of an individual's N budget could be supported by direct uptake of organic

relative to inorganic N (Chapin, Moilanen, and Kielland, 1993b; Kielland, 1994, 1997).

Subsequent studies have shown that plants from a wide range of ecosystems can directly

access organic N as an important part of their N nutrition (Lipson and Nasholm, 2001).

Few studies, however, have examined the ability of plants from tropical wet forests to









take up organic nitrogen and to our knowledge no studies have examined this ability in

the ferns.

The ferns pose a unique set of ecological constraints due to both the dispersal of

tiny wind-blown spores and the occurance of an independent and free-living haploid

gametophyte. The mineral nutrition of fern sporophytes is not well studied; however,

evidence indicates that fern sporophytes behave as seed plant sporophytes when

confronted with increased inorganic N (Prange and Ormrod, 1982; Walker and Aplet,

1994; Pillai and Ong, 1999). Less is known of the mineral nutrition of gametophytes and

unlike sporophytes (which rely on well developed root systems), fern gametophytes are

thought to rely primarily on rhizoid uptake of nutrients with possible uptake of water and

nutrients across the thallus (Racusen, 2002). The ability of fern gametophytes to grow on

different N forms was reviewed by Miller (1968), and there is limited evidence to show

that the gametophytes of at least one species can grow well on specific mixtures of amino

acids in the absence of inorganic forms.

For plants other than ferns, changes in 815N values have been shown to occur

through ontogeny and with increases in plant size (Zotz, 1997; Schmidt, Stuntz, and Zotz,

2001; Hietz and Wanek, 2003; Reich et al., 2003; Zotz et al., 2004; Casper, Forseth, and

Wait, 2005). In the case of hemiepiphytic plants, changes occur as a direct result of the

connection of once aerial roots to terrestrial water and nutrient pools (Putz and Holbrook,

1989; Field, Lawton, and Dawson, 1996; Wanek et al., 2002). As ferns continue their

life-cycle from gametophyte to sporophyte, radical changes in their ecophysiology,

especially nutrient relations are likely to occur.









The goal of this paper is to determine if species differ in their ability to take up

different forms of nitrogen in both the gametophyte and sporophyte generations and to tie

this to the natural abundance of 615N to determine if species access different nitrogen

source pools. We then compare these data across epiphytic and terrestrial species and a

developmental series of a hemiepiphytic species to better understand the dynamics of

nitrogen nutrition of different life forms.

Material and Methods

Study Site

This study was conducted at La Selva Biological Station of the Organization for

Tropical Studies in Heredia Province, Costa Rica (10026' N, 84o00' W). La Selva is a

1400 ha tropical wet-forest positioned in the Caribbean lowlands with an average

monthly temperature of 25.8 C and annual rainfall of 4000 mm per year (Sanford et al.,

1994). The site boasts a diversity of ferns with multiple species from epiphytic,

hemiepiphytic and epiphytic life forms (Grayum and Churchill, 1987).

Study Species

The gametophytes and sporophytes of 10 species were field collected from

100X100m grids to control for differences in soil type in the terrestrial species and from

the trees in the case of epiphytic species (Table 4-1). The following species were

sampled:

Adiantum latifolium Lam. is a terrestrial species that is common in disturbed areas

in both primary and secondary forests. The species typically grows along trail sides and

can be encountered under a wide rage of light and soil regimes.

Danaea nodosa (L.) Sm. and Danaea wendlandii Rchb. F. are both eusporangiate

ferns and as such have gametophytes that are several cell layers thick. Danaea









wendlandii is perhaps the most common fern at La Selva and often grows on upland well

drained sites and in disturbed understory habitats. Danaea nodosa has similar habitat, but

is more often associated with wetter sites becoming most abundant along creek sides.

Diplazium subsilvaticum H. Christ is a terrestrial arborescent species that is

common along creek banks and wetter areas in both primary and secondary forests.

Lomariopsisjapurensis (Mart.) J. Sm. and Lomariopsis vestita E. Foum are both

understory hemiepiphtyes. In both species, the gametophytes develop on the trunks of

small trees and remain epiphytic throughout their life. Young sporophytes produce roots

that grow down and contact the soil and rely on the host tree for support. Adult plants

never loose contact with the forest floor.

Olfersia cervina (L.) Junze has been classified as both a terrestrial and

hemiepiphyte species. It is restricted to grow on soils with high organic content and is

most commonly found growing on rotting logs, but can also grow on large tree trunks

where sufficient detritus has accumulated.

Elaphoglossum latifolium (Sw.) J. Sm. and Campyloneurum brevifolium (Lodd.

Ex Link) Link are both canopy epiphytes with the former more often found in highly

exposed portions of the canopy and often on bare bark. The latter species also occurs in

exposed sites, but is most common in the inner canopy rooted in canopy soil organic

matter.

Antrophyum lineatum (Sw.) Kaulf: An understory epiphyte that grows on the

trunks of living trees. Both gametophytes and sporophytes are common in primary and

secondary forests at La Selva.









Isotopic Natural Abundance and 61 N Labeled Uptake

Foliar and gametophytic samples for natural abundance of all nine species were

field collected within a 3-wk period during June 2005. Collections were brought back to

the lab where they were washed in deionized water to remove all soil and then dried at

60C for 48h. Samples were analyzed for nitrogen concentration and isotope ratio using a

Costech elemental analyzer coupled with a Finnigan Delta XL PlusTM continuous flow

mass spectrometer at the University of Florida, Gainesville. Based on repeat analyses of

NIST peach leaves standard (SRM 1547; 615N 1.91%o), average Is precision was 0.07%o

for 615N.

In order to compare uptake of both organic and inorganic N forms we used

excised root techniques (Treseder and Vitousek, 2001) in sporophtyes and whole plant

uptake in gametophytes of Danaea wendlandii, Lomariopsis vestita, and Campyloneurum

brevifolium. Immediately prior to the uptake trials, plants were field collected and

brought back to the lab where roots or gametophytes were rinsed with deionized water to

remove soil and other debris. For sporophytes, fine roots were selected, excised, and

placed into a series of solutions containing increasing concentrations of 615N labeled

organic and inorganic N forms (see below). For gametophyte trials, 2-5 individual

gametophytes of similar size and maturity were placed directly into a separate set of

solutions. All samples were allowed to incubate for 60min in solutions containing 0

deionizedd water only), 10, 50, 100, 300, and 500 imol concentrations containing only

labeled NH4+, NO3-, (99 atom%), a cocktail of equal proportions of the amino acids:

Aspartic and Glutamic Acids, and Glycine (98 atom%), and a cocktail of all solutions

(NH4 + NO3- + the 3 amino acids). All solutions were amended with 0.01 mol/L sucrose

as an energy source and 0.5 mmol/L CaC12 to maintain membrane integrity (Kielland









1997). Immediately following the incubation trial, roots and gametophytes were removed

and rinsed for 2min in a solution containing Immol/L KC1 to remove any excess 615N

from the external surfaces. The material was then dried at 60C for 73h, weighed, and

ground for analysis of 615N using the same Finnigan Delta XL PlusTM continuous flow

mass spectrometer.

Nutrient Uptake Calculations

815N enrichment was calculated as F=[T(As-AB)]/AF, where F is the weight of N

derived from the 615N tracer, Tis the total weight of N in the sample, As is atom% excess

615N in the labeled sample, AB is atom% excess 615N in the natural abundance sample,

and AF is atom% excess in the 615N tracer (Knowles and Blackburn, 1993). To calculate

the kinetic uptake parameters of maximum uptake (Vmax) and the saturation constant

(Km), we fitted the data to a Michaelis-Menten function. We also quantified the .mols of

N per g root dry mass against the solution concentration using V= Vmax S/ Km + S, where

V is the velocity of uptake, and S is the concentration. The parameter Vmax is an estimate

of the maximum uptake rate for a given ion and is controlled by the activity of membrane

bound proteins specific to that ion. The value of experimental calculations of Vmax is that

it gives an estimate of the root's/gametophyte's total capacity of ion uptake. The value

Km is estimated to describe the affinity of specific membrane-bound proteins to a given

ion and is related to the capacity to utilize low concentrations of this ion. Roots with

lower Km values have higher affinities at low concentrations for the ion in question.

Results

61sN Natural Abundance and N concentration (mg g-)

Species differed significantly in 615N and N concentration (mg g-l) values (F=3.06,

p=0.0053 and F=3.69, p=0.0009 respectively). There were also significant differences









within and between life forms (615N F=3.74, p=0.029; N concentration (mg g-') F=10.66,

p<0.0001), generations (615N F=15.52, p=0.0002; N concentration (mg g-1) F=57.65,

p<0.0001), and a significant life form by generation interaction (615N F=16.95, p<0.0001;

N concentration (mg g-l) F=8.85, 0.0004) (Fig. 4-1, 4-2a). Within both hemiepiphytic and

terrestrial species, the 615N values of gametophytes were more enriched F=64.93,

p<0.0001 and F=3.84, p=0.0061) than that of sporophtyes; whereas, the opposite was the

case for the epiphytes (F=3.99. p=0.059). Across life forms, the gametophytes of

terrestrial species were slightly depleted yet not significantly so when compared to

epiphytic and hemiepiphytic species (F=2.63, p=0.086). Greatest differentials were

observed among the sporophtyes with hemiepiphytes significantly more depleted than

terrestrial species than epiphytic species (F=17.12, p<0.0001). Gametophytes exhibited

significantly higher N concentration (mg g-1) relative to sporophytes within each life form

and varied between life forms with epiphytes and hemiepiphytes exhibiting higher N

concentration (mg g-1) in gametophytes and sporophtyes than terrestrial species (Fig. 4-

2b). To better understand ontogenetic shifts that occur in hemiepiphytic ferns, we

measured variation in 15N natural abundance of different stages of Lomariopsis vestita.

There was a clear series of increasing 615N enrichment from epiphytic gametophytes and

sporophtyes to terrestrially-rooted adult sporophtyes (Fig. 4-3).

61sN Labeled Uptake

The gametophytes and sporophytes of both Danaea wendlandii and

Campyloneurum brevifolium exhibited uptake capacity of both inorganic and organic

forms of N. For both species uptake of NO3- in both gametophytes and sporophytes was

limited. The gametophytes of both species had higher Vmax for all N forms than

sporophytes. The gametophytes and sporophytes of C. brevifolium had higher Vmax









values for the amino acid mix followed by NH4 and the all solution cocktail (Fig. 4-

4a&b and Fig. 4-5a&b) and all solution cocktails. The gametophytes ofD. wendlandii

had highest Vmax values for the all solution cocktail where the Vmax for NH4+ and amino

acid was similar (Fig. 4-5c). Km values exhibited considerable variation both within and

between species and generations. In all cases, Km was lowest and therefore affinity

highest for NO3- relative to all other N forms (Fig. 4-6a-d).

Discussion

The results from the uptake studies indicate that ferns show preference for

specific N forms and that they do so differently in sporophyte and gametophyte

generations. In the case of the epiphytic C. brevifolium, both gametophytes and

sporophtyes exhibited high potential for uptake of amino acid N followed by inorganic

NH4+ (Fig. 4-5a-b). Uptake potentials shifted slightly in the gametophytes of the

terrestrial D. wendlandii with uptake of amino acids and NH4+ essentially equal. The

gametophytes and sporophytes of the epiphytic species exhibited higher uptake capacities

for N-derived from amino acids relative to the terrestrial Danaea wendlandii.

Campyloneurum brevifolium is a mid- to low-canopy species that is frequently rooted in

canopy soil; whereas, Danaea wendlandii is a species that is always rooted on mineral

soil. The occurrence of species on such different soil types is likely to produce radically

different nitrogen nutrition and produce different nutrient use strategies between such

species.

Canopy soil is fundamentally different from terrestrial soil in that the former is

almost entirely organic. In a study on soil nutrients at La Selva, Cardelus and Mack (pers.

com.) have shown that canopy soil organic matter had significantly greater bulk nitrogen,

NH4 and dissolved organic nitrogen than terrestrial forest floor soils and that nitrogen









mineralization rates were significantly lower in the canopy. They were also able to show

that within canopy soils, the concentration of NO3 was low and NH4+ and dissolved

organic nitrogen dominate this matrix. For this reason, it is not surprising that epiphytic

species would exhibit preferential uptake ofNH4+ and organic nitrogen.

Both the uptake rate (Vmax) and half saturation constants (Km) were higher for

NH4+ than either amino acids or NO3s for the sporophytes ofDanaea wendlandii (Fig 4-5,

4-6). In the gametophytes of this species, the values of Km for amino acids were several

fold higher than those from the sporophtyes and from the gametophytes and sporophtyes

of Campyloneurum brevifolium. Uptake rates of amino acids and NH4+ from these same

gametophytes were not significantly different. These patterns indicate that NH4+ is an

important N source for both, but especially, terrestrial sporophtyes in this study. NO3s

concentration within terrestrial soils at this site is high (Cardelus and Mack pers. com.).

Nitrate is highly mobile and ferns must compete with microbes and other plants for this

resource and may have partitioned uptake to NH4 and amino acids to avoid or lessen this

competition. Such plants would not be highly invested in NO3s carriers and would be

expected to exhibit higher affinities for this ion at lower concentrations; a result

demonstrated by this study (Fig. 4-6).

The importance of amino acids as components of species' N budgets is receiving

increased attention and has been shown in species from tundra (Kielland, 1994), to

temperate (Finzi and Berthrong, 2005) and subtropical (Schmidt and Stewart, 1999)

ecosystems. We believe that our data are some of the first to demonstrate this ability in

species from tropical lowland forests and clearly the first to do so in the ferns. The

significance of amino acids varies but clearly makes up a critical component (>50% by









some estimates) of species' N budgets from tundra ecosystems (Kielland 1994). In many

ways, canopy soil is functionally equivalent to tundra and boreal soils as it is organic in

origin and potentially has high concentrations of free amino acids. As such, there may be

convergence to greater investment in amino acid uptake in organic soils.

The gametophytes of both species had higher uptake rates of all N forms

excluding NO3- relative to sporophytes. There are fundamental differences in anatomy

and morphology of these two stages of the life cycle and comparisons across stage must

be made with care. Gametophytes produce primitive yet functional rhizoids that likely aid

in nutrient uptake (Smith, 1972a, 1972b), yet they may also take in nutrients via diffusion

across cells of the thallus (Racusen, 2002). This could result in much greater

gametophyte surface area and transporter density compared to root surface area and result

in greater uptake per unit mass in gametophytes. Expression of uptake rates on an area

basis is made difficult as the gametophytes of both species develop complex three

dimensional morphologies that make accurate determination of area difficult.

There are a number of mechanisms that control the natural abundance 615N

signatures in plant tissues and for this reason, 615N signatures reflect a series of integrated

fractionation events (Evans, 2001; Robinson, 2001; Dawson et al., 2002). In spite of the

complexities of interpreting natural abundance 615N signatures, such data provide

evidence of a process when confounding variables that influence fractionation can be

identified or eliminated (Robinson, 2001). Differences in plant 615N signatures have been

shown to be primarily related to 1) uptake of different N sources with distinct signatures

(Robinson, 2001), 2) N availability and plant demand (Kolb and Evans, 2003), 3)









mycorrhizal associations (Hobbie and Hobbie inpress), 4) rooting depth (Nadelhoffer

and Fry, 1994) or a combination of these events (Dawson et al., 2002).

Both N availability and plant demand can have impacts on tissue 615N values

through kinetic fractionations Kolb and Evans (2003) have shown that when N supply is

greater that a plant's assimilatory ability, plant tissues can become highly depleted in 15N

relative to the source. This, however, only occurs in ecosystems where external N

concentrations are high and such kinetic fraction seems unlikely to be responsible for the

differences observed in our study. It is possible that mechanisms other than supply and

demand drive kinetic fractionations differently in gametophytes and sporophtyes.

Unfortunately, little is known of how root vs. rhizoid/thallus uptake differentially

fractionate N.

Ectomycorrhizal associations can result in large fractionation events (8-10 /oo)

whereas fractionation caused by arbuscular mycorrhizal symbioses seems to have little

effect on 615N values of host plants (Schmidt and Stewart, 2003). Both terrestrial and

epiphytic leptosporangiate fern species have arbuscular mycorrhizal symbioses in the

sporophtyes, but never in the gametophytes (Gemma, Koske, and Flynn, 1992). The

differences in natural abundance 615N signatures between gametophytes and sporophtyes

are unlikely due to such associations.

Plant uptake from source pools with different 615N signatures is a major

contributor influencing tissue values. Plants can take up both organic and inorganic

nitrogen and each of these sources has different signatures: NO3 is highly depleted

relative to NH4+ which can be more depleted that amino acids. In our data, gametophyte

natural abundance 615N signatures are either significantly depleted or equal to sporophyte









values. The data from the uptake experiment indicate that NO3 is not a major source of

either gametophyte or sporophtyes N nutrition and that source alone is not responsible for

the differences between gametophyte and sporophtyes. Sources can however be derived

from N species from enriched soil or highly depleted atmospheric sources. Evidence

indicates that epiphytes rely heavily on depleted atmospheric N sources (Hietz et al.,

2002) and the result is major offset of 15N values when compared to terrestrially rooted

plants (Watkins, unpublished data). This may also help explain the differences between

gametophytes and sporophtyes. As throughfall leaches through the canopy to the forest

floor, it may contain significant concentrations of depleted N (Cardelus and Mack pers.

com.) which may then be taken up more directly by gametophytes. Canopy epiphytic

gametophytes also rely on depleted N species and may exhibit greater direct atmospheric

uptake than sporophytes. Our uptake data indicate that overall uptake rate in

gametophytes is much greater than that in sporophtyes. This may provide gametophytes

greater opportunity for uptake of depleted pools that may be rapidly diluted by rainfall.

Several studies have now shown that tissue 615N values from deep rooted

individuals are more enriched relative to shallow rooted individuals (Nadelhoffer and

Fry, 1988; Nadelhoffer and Fry, 1994; Handley and Scrimgeour, 1997). Rooting depth is

a potentially major contributor to the observed differences in 615N values between the

surface rooted gametophytes and deeper sporophyte roots.

One way to observe shifts in N sources is to follow ontogenetic series and track

changes in 615N values. The gametophytes of Lomariopsis vestita are epiphytes on

understory trees and produce epiphytic sporophtyes with roots that grow down the trunk

into the soil. The difference between gametophytes and young sporophtyes that were not









attached to the soil were large with the latter being more enriched (Fig. 4-3). This is

possibly due to a combination of atmospheric uptake in gametophytes and quickly shifts

to direct root uptake in sporophtyes that may be more efficient in tapping nutrients from

more enriched soil pools. Terrestrially rooted young and adult individuals would be

predicted to exhibit enriched 815N signatures relative to epiphytic individuals. Such

plastic abilities are critical in the life of hemiepiphytes that face frequent water stress and

a temporally heterogeneous nutrient environment.

Conclusions

The observed differences in foliar 815N values and evidence of differential uptake

of N forms indicate that ferns can partition N by form. The preference of N form varied

with greater preference on organic N in the epiphytic vs. terrestrial species and between

gametophytes and sporophtyes. These data indicate that there are different nutrient use

strategies between the two life forms and between generations. Individuals that can

circumvent mineralization in the N cycle by direct amino acid uptake may have a

tremendous competitive advantage relative to those relying on inorganic forms. What is

critically needed are studies that incorporate dual labeled C and N isotopes to precisely

determine if amino acids are hydrolyzed at the cell membrane or if they are taken directly

into plants.













Species
Adiantum latifolium Lam.
Danaea nodosa (L.) Sm.
Danaea wendlandii Rchb. F.
Diplazium subsilvaticum H. Christ
Lomariopsisjapurensis (Mart.) J. Sm.
Lomariopsis vestita E. Foum
Olfersia cervina (L.) Junze
Elaphoglossum latifolium (Sw.) J. Sm.
Campyloneurum brevifolium (Lodd. Ex Link)
Antrophyum lineatum (Sw.) Kaulf


Life Form
Terrestrial
Terrestrial
Terrestrial
Terrestrial
Hemiepiphyte
Hemiepiphyte
Hemiepiphyte-Terrestrial
Epiphyte
Epiphyte
Epiphyte


Distribution
Disturbed areas, high medium light
Understory, low light
Understory, disturbed low light areas
Understory exposed mesic areas
Understory secondary and primary forests
Understory secondary and primary forests
Understory, on mounds of organic matter or decaying trees
Highly exposed, on bare bark in canopy
Moderate light, rooted in soil, inner canopy
Understory secondary and primary forests


Table 4-1. Species, life form and ecology for the natural abundance and uptake experiments






































Danea nodosa







F=6.803, p=0.0798


Olfersia cervina








F=11.77, p=0.041

Lomariopsis vestita








F=51.40, p<0.0001

Gametophyte Sporophyte


F=4.263, p=0.178

Microgramma reptans








F=33.01, p=0.0045

Gametophyte Sporophyte


Figure 4-1. Sporophtyic and Gametophytic 615N natural abundance signatures of 10
tropical fern species. Tissue was field collected from 100X100m grids to
control for differences in soil type in the terrestrial species and from the same
trees in the case of epiphytic species


Adiantum latifolium








F=2.131, p=0.218


Lomariopsis jauperensis








F=7.73, p=0.49


Danea wendlandii





7, T r


F=24.53, p=0.0006


Polytenium








F=0.838, p=0.383


0

-2

6

4
z
L 2

0

-2

6

4


Campylonerum brevifolium








F=0.946, p=0.349


Elaphoglossum latifolium































Epiphyte Hemi-Epiphyte Terrestrial
Epiphyte Hemi-Epiphyte Terrestrial


Aab


1 -


0


-1 -




5


4


3
E
2


1


0


Terrestrial


Figure 4-2. Sporophtyic and Gametophytic (a) 615N natural abundance signatures and (b)
N concentration (mg g-) of epiphytic, terrestrial, and hemiepiphytic tropical
fern species. Post hoc test were generated using Tukey tests. Capitol letters
refer to within life-form whereas lower case letter refer to across life form
comparisons


S Gametophyte Bb
m Sporophyte




Bc



Aa
Aa

T Ba Aa


Epiphyte Hemi-Epiphyte







71




5
F=20.34, p<0.0001
C
4T


3
BC
z
Ln 2
-a AB



A
0


-1 i i
Gametophytes Young Young Adult
Sporophytes Sporophytes Sporophytes
Not Attached Attached Attached

Figure 4-3. 615N natural abundance signatures from the hemiepiphytic fern Lomariopsis
vestita. Gametophytes of this species are completely epiphytic on understory
trees. Young sporophtyes are initially produced that have true roots, but no
connection to the forest floor at very early stages. Young sporophtyes
eventually attach to the soil and rooting depth increases with adult plants











160
S140
120
5 100
E
= 80
60
rD 40
20
0


0 100 200 300 400 500


N Concentration (umol/L)


C DW Gametophyte











0 100 200 300 400 500
N Concentration (umol/L)


B CB Sporophyte










0 100 200 300 400 500


N Concentration (umol/L)


D DW Sporophyte










0 100 200 300 400 500
N Concentration (umol/L)


-- Amino Acid Mix
-- All Solutions
--- NH4
N03


Figure 4-4. Uptake curves from 615N labeled solutions. Fine roots were selected, excised,
and placed into a series of solutions containing increasing concentrations of
815N labeled organic and inorganic N forms. For the gametophyte trials
individual gametophytes of similar size and maturity were placed directly into
a separate set of solutions. All samples were allowed to incubate for 60min in
solutions containing only 15N labeled NH4 NO3-, (99 atom%), a cocktail of
equal proportions of the amino acids: Aspartic and Glutamic Acids, and
Glycine (98 atom%), and a cocktail of all solutions (NH4+ + NO3- + the 3
amino acids)


A CB Gametophyte


160
S140
- 120
- 100
E 80
S60
rQ 40
1 20
_ 0












Campyloneurum brevifolium
Gametophyte


200




-150
_c


0)
S100
-o
E


>50




0







200




,150
_c


0)
r--


z 100
-6
o
E

E
50




0


200




-150


-o
0)
z 100
Z-6O
o
E


50


Campyloneurum brevifolium
Sporophyte


NH4 N03 AA
Nitrogen Form


200




,150
_c
-o

0)
z 100
o
E


50




0


NH4 N03 AA
Nitrogen Form


Danea wendlandii
Sporophyte

D F=61.55, p<0.0001












A
-- A


B
B B


NH4 N03 AA All
Nitrogen Form


Figure 4-5. Uptake saturation values (Vmax) of each N form derived from Michaelis-
Menten functions of the data from Fig. 4-2, error bars are standard errors.
Campyloneurum brevifolium is a mid-canopy epiphyte of exposed habitats;
Danaea wendlandii is a low-light understory terrestrial species


F=83.34, p<0.0001


-




-




-




-


NH4 N03 AA All
Nitrogen Form


Danea wendlandii
Gametophyte

C F=13.0, p=0.0019




C


A




- ^




B

.__L -- -


B
F--














Campyloneurum brevifolium
Gametophyte


Campyloneurum brevifolium
Sporophyte


NH4 N03 AA All NH4 N03 AA All
Nitrogen Form Nitrogen Form


Danea wendlandii
Gametophyte


NH4 N03 AA
Nitrogen Form


-j
-0
E 150
E
100


50


NH4 N03 AA
Nitrogen Form


Figure 4-6. V2 Uptake saturation values (Km) of each N form derived from Michaelis-
Menten functions of the data from Fig. 4-2, error bars are standard errors.
Campyloneurum brevifolium is a mid-canopy epiphyte of exposed habitats;
Danaea wendlandii is a low-light understory terrestrial species.


Danea wendlandii
Sporophyte













CHAPTER 5
CONCLUSIONS

This dissertation is one of the first attempts to examine and apply modern

ecological and ecophysiological techniques to the study of fern gametophyte ecology.

This work has demonstrated that fern gametophytes can be extremely long-lived in situ

and that there are differences in factors that influence the distribution and demography of

epiphytic and terrestrial ferns. Differences in life history and the way that epiphytic and

terrestrial life-forms respond to disturbance and light provide evidence for adaptively

meaningful variation in life histories that has evolved in the two groups. Epiphytic

species have evolved in a high light, highly competitive, yet relatively stable matrix. Such

environments reduce the light limitations encountered by terrestrial species, yet they

incorporate closer contact with bryophytes. These habitat mediated conditions may be

largely responsible for the observed variation in longevity.


Dassler and Farrar (1997) have argued that differences in gametophyte longevity

between epiphytic and terrestrial species have largely evolved due to pressures from the

genetic consequences of intergametophytic selling. Asexually reproducing indeterminate

gametophytes of many epiphytes can produce large and long-lived clones. Such clones

greatly increase the longevity of individual genotypes which is hypothesized to increase

the chance of outcrossing. The data generated from chapter one clearly show that

epiphytic gametophytes are significantly longer lived than terrestrial species. One

remarkable discovery is that epiphytic gametophytes can live for years where even









terrestrial species on relatively stable substrates rarely lived beyond 6 months. I have

observed gametophytes of some understory epiphytes that are over 6 years old.


Are the differences in longevity between epiphytes and terrestrial species related to

some adaptively meaningful variation between the life-forms as has been hypothesized,

or is it simply a result of some intrinsic instability of epiphytic vs. terrestrial habitats?

The answer to this question remains elusive as there is no clear understanding of the

differences in disturbance between epiphytic and terrestrial habitats. The data in my study

would indicate that epiphytic habitats are far more stable than terrestrial habitats. This

clearly needs to be established and a much greater survey of demography needs to be

undertaken to better understand the extent of the differences that I have reported.

Gametophytes that can live for months and especially those that live for years have

to cope with stress associated with extreme abiotic variation. The second part of my

dissertation has revealed a surprising and completely unexpected degree of extreme

gametophyte stress tolerance to desiccation across several species. All species surveyed

exhibited more desiccation tolerance than current pteridological dogma would suggest. In

addition, such tolerance was clearly linked to species sporophyte ecology with those from

drought prone habitats, such as the epiphytes in the study, exhibiting greater degrees of

tolerance compared to those in mesic habitats. Epiphytic species were also robust in

dealing single dry down events and exhibited significantly greater recovery following

extreme desiccation intensities and multiple desiccation cycles compared to more mesic

terrestrial species. Not only are epiphytic species longer-lived, they are also considerably

more desiccation tolerant: two characters that are likely connected.









Stress tolerance has been show to structure some bryophyte communities (Cleavitt,

2002) but I have been unable to find any additional work to suggest that ferns are sorting

along dines of stress tolerance. Much additional work needs to be completed on

gametophyte physiology to truly understand the role that gametophytic longevity and

desiccation tolerance plays in sorting species. Additional studies need to include

combinations of temperature and light stress with desiccation to develop a better

understanding of the interaction of these characters and the relative tolerance of more

species. This work also has basic science applications beyond ferns and can be applied to

aspects directly related to genetic engineering of desiccation tolerance in crop plants.

One factor that seems closely tied to sorting of terrestrial fern sporophtyes are

edaphic factors (Tuomisto, 1998, Tuomisto, 1994, Tuomisto, 2002). The role that

nutrients play in shaping fern gametophyte distributions is virtually unknown. My work

on nitrogen relations of tropical fern gametophytes has revealed unexpected versatility in

nitrogen acquisition between both gametophytes and sporophtyes and between epiphytic

and terrestrial species. Of great significance was the discovery that ferns can partition

nitrogen by form and have the ability to take up amino acids and use them as an

important component of their nitrogen budgets. The importance of amino acid uptake as

critical components of species' N budgets is currently receiving increased attention and

has been shown in species from tundra (Kielland, 1994), to temperate (Finzi and

Berthrong, 2005) and subtropical (Schmidt and Stewart, 1999) ecosystems. Nitrogen

associated with amino acids is clearly important in both the N cycle of tropical fern

gametophytes and sporophtyes and such flexibility in accessing different nitrogen forms

may provide species with differential competitive abilities result in one mechanism by






78


which species are sorted along nutrient gradients. What is critically needed are studies

that incorporate dual labeled C and N isotopes to precisely determine if amino acids are

hydrolyzed at the cell membrane or if they are taken directly into plan.


















LIST OF REFERENCES


ALPERT, P. 2000. The discovery, scope, and puzzle of desiccation tolerance in plants.
Plant Ecology 151: 5-17.

2005. The limits and frontiers of desiccation-tolerant life. Integrative and
Comparative Biology 45: 685-695.

ALPERT, P., and M. J. OLIVER. 2002. Drying without dying. In M. Black and H. W.
Pritchard [eds.], Desiccation and Survival in Plants, 3-43. CABI Publishing,
Oxon.

ATKINSON, L. R.,and A. G. STOKEY. 1964. Comparative morphology of the gametophytes
of homosporous ferns. Phytomorphology 14: 51-70.

BEWLEY, J. D. 1979. Physiological-aspects of desiccation tolerance. Annual Review of
Plant Physiology and Plant Molecular Biology 30: 195-238.

BOLD, H. C. 1957. Morphology of Plants. Harper and Row, New York.

BOWER, F. O. 1923. The Ferns. Cambridge University Press, Cambridge.

CARDELUS, C. L., and R. CHAZDON. 2005. Inner-crown microenvironments of two
emergent tree species in a lowland wet forest. Biotropica 37: 238-244.

CASPER, B. B., I. N. FORSETH, and D. A. WAIT. 2005. Variation in carbon isotope
discrimination in relation to plant performance in a natural population of
Cryptantha flava. Oecologia 145: 541-548.

CHAPIN, F. S., L. MOILANEN, and K. KIELLAND. 1993a. Preferential use Of organic
nitrogen for growth by a nonmycorrhizal arctic sedge. Nature 361: 150-153.

1993b. Preferential Use of Organic Nitrogen for Growth by a Non-Mycorrhizal
Arctic Sedge. Nature 361: 150-153.

CHIOU, W.-L., and D. R. FARRAR. 1997. Comparative gametophyte morphology of
selected species of the family Polypodiaceae. American Fern Journal 87: 77-86.









CLEAVITT, N. L. 2002. Stress tolerance of rare and common moss species in relation to
their occupied environments and asexual dispersal potential. Journal of Ecology
90: 785-795.

COUSENS, M. I. 1979. Gametophyte ontogeny, sex expression, and genetic load as
measures of population divergence in Blechnum spicant. American Journal of
Botany 66.

1981. Blechnum spicant: habitat and vigor of optimal, marginal, and disjunct
populations and field observations of gametophytes. Botanical Gazette 142: 251-
258.

1988. Reproductive strategies of pteridophytes. In J. L. Doust and L. L. Doust
[eds.], Plant Reproductive Ecology: Patterns and Strategies, 307-328. Oxford
University Press, New Youk.

COUSENS, M. I., D. G. LACEY, and E. M. KELLY. 1985. Life-History Studies Of Ferns A
Consideration Of Perspective. Proceedings Of The Royal Society Of Edinburgh
Section B-Biological Sciences 86: 371-380.

COUSENS, M. I., D. G. LACEY, and J. M. SCHELLER. 1988. Safe sites and the ecological
life history ofLorinseria areolata. American Journal ofBotany 76: 797-807.

CRIST, K. C., and D. R. FARRAR. 1983. Genetic load and long-distance dispersal in
Aspleniumplatyneuron. Can. J Bot. 61: 1809-1814.

CSINTALAN, Z., M. C. F. PROCTOR, and Z. TUBA. 1999. Chlorophyll fluorescence during
drying and rehydration in the mosses Rhytidiadelphus loreus (Hedw.) Warnst.,
Anomodon viticulosus (Hedw.) Hook. & Tayl. and Grimmia pulvinata (Hedw.)
Sm. Annals ofBotany 84: 235-244.

DASSLER, C. L., and D. R. FARRAR. 1997. Significance of form in fern gametophytes:
clonal, gemmiferous gametophytes of Callistopteris baueriana
(Hymenophyllaceae). International journal of plant sciences 158: 622-639.

2001. Significance of gametophyte form in long-distance colonization by
tropical, epiphytic ferns. Brittonia 53: 352-369.

DAWSON, T. E., S. MAMBELLI, A. H. PLAMBOECK, P. H. TEMPLER, and K. P. Tu. 2002.
Stable isotopes in plant ecology. AnnualReview of Ecology and Systematics 33:
507-559.

DELTORO, V. I., A. CALATAYUND, G. GIMENO, and E. BARRENO. 1998. Water relations,
chlorophyll fluorescence, and membrane permeability during desiccation in
bryophytes from xeric, mesic and hydric environments. Canadian Journal of
Botany 76: 1923-1929.









DEMMIG-ADAMS, B., and W. W. ADAMS. 1992. Photoprotection and Other Responses of
Plants to High Light Stress. Annual Review ofPlant Physiology and Plant
Molecular Biology 43: 599-626.

DEMMIG-ADAMS, B., and W. W. ADAMS III. 1992. Photoprotection and other responses of
plants to high light stress. Annu. Rev. Plant Physiol. Plant Mol. Biol. 43: 599-626.

EVANS, R. D. 2001. Physiological mechanisms influencing plant nitrogen isotope
composition. Trends in Plant Science 6: 121-126.

FARRAR, D. R. 1967. Gametophytes of four tropical fern genera reproducing
independently of their sporophytes in the southern Appalachians. Science 155:
1266-1267.

1971. The biology of ferns with asexually reproducing gametophytes in the
eastern United States. Ph.D. Dissertation. Ph.D. Dissertation, University of
Michigan, Ann Arbor, MI.

1990. Species and evoluiton in asexually reproducing independent fern
gametophytes. Systematic Botany 15: 98-111.

1998. The tropical flora of rockhouse cliff formations in the eastern United
States. Journal of the Torrey Botanical Club 125: 91-108.

FIELD, T. S., R. O. LAWTON, and T. E. DAWSON. 1996. Comparative nutrient relations in
canopy-rooted and ground-rooted Didymopanax pittieri Marchal (Araliaceae)
hemiepiphytes in a wind-exposed tropical montane rain forest. Biotropica 28:
774-776.

FINZI, A. C., and S. T. BERTHRONG. 2005. The uptake of amino acids by microbes and
trees in three cold-temperate forests. Ecology 86: 3345-3353.

FRAZER, G. W., C. D. CANHAM, P. SALLAWAY, and D. MARINAKIS. 1999. Gap Light
Analyzer. Simon Fraser University & Institute for Ecosystem Studies, Burnaby,
British Columbia, Canada & Millbrook, New York.

GAFF, D. 1997. Mechanisms of desiccation tolerance in resurrection plants. In A. S. Basra
and R. K. Basra [eds.], Mechanisms of environmental stress resistance in plants,
43-58. Harwood Academic Publishers, London.

GAFF, D. F. 1987. Desiccation tolerant plants in South America. Oecologia (Berlin) 74:
133-136.

GEMMA, J. N., R. E. KOSKE, and T. FLYNN. 1992. Mycorrhizae in Hawaiian
pteridophytes: occurrence and evolutionary significance. American Journal of
Botany 79: 843-852.










GILBERT, O. L. 1970. Biological flora of the British Isles. Dryopteris villarii (Bellardi)
Woynar (Aspidium rigidum.; Lastrea rigidia (Sw.) C. Presl).

GRAYUM, M. H., and H. W. CHURCHILL. 1987. An introduction to the Pteridophyte flora
of Finca La Selva, Costa Rica. American Fern Journal 77: 73-89.

GREER, G. K., and B. C. MCCARTHY. 1999. Gametophytic plasticity among four species
of ferns with contrasting ecological distributions. International Journal of Plant
Sciences 160: 879-886.

HANDLEY, L. L., and C. M. SCRIMGEOUR. 1997. Terrestrial plant ecology and N-15
natural abundance: The present limits to interpretation for uncultivated systems
with original data from a Scottish old field, Advances in Ecological Research, Vol
27, 133-212.

HARTEN, J. B., and W. G. EICKMEIER. 1987. Comparative Desiccation Tolerance of 3
Desert Pteridophytes Response to Long-Term Desiccation. American Midland
Naturalist 118: 337-347.

HIETZ, P., and 0. BRIONES. 1998a. Correlation between water relations and within-
canopy distribution of epipytic ferns in a Mexican cloud forest. Oecologia 114:
305-316.

1998b. Correlation between water relations and within-canopy distribution of
epiphytic ferns in a Mexican cloud forest. Oecologia 114: 305-316.

HIETZ, P., and W. WANEK. 2003. Size-dependent variation of carbon and nitrogen isotope
abundances in epiphytic bromeliads. Plant Biology 5: 137-142.

HIETZ, P., W. WANEK, R. WANIA, and N. M. NADKARNI. 2002. Nitrogen-15 natural
abundance in a montane cloud forest canopy as an indicator of nitrogen cycling
and epiphyte nutrition. Oecologia 131: 350-355.

HOLBROOK-WALKER, S. G., and R. M. LLOYD. 1973. Reproductive biology and
gametophyte morphology of the Hawaiian fern genus Sadleria (Blechnaceae)
relative to habitat diversity and propensity for colonization. Botanical Journal of
the Linnean Society 67: 157-174.

HOLLANDER, M., and D. A. WOLF. 1999. Nonparametric statistical methods. John Wiley
and Sons Inc., New York.

HOLTTUM, R. E. 1938. The ecology of tropical pteridophytes. In F. Verdoorn [ed.],
Manual of pteridology., 433. M. Nijhoff, The Hague.









HORTON, P., A. V. RUBAN, and R. G. WALTERS. 1996. Regulation of light harvesting in
green plants. Annual Review OfPlant Physiology And Plant Molecular Biology
47: 655-684.

JONES, M. M., H. TUOMISTO, D. B. CLARK, and P. OLIVAS. 2006. Effects of mesoscale
environmental heterogeneity and dispersal limitation on floristic variation in rain
forest ferns. Journal of Ecology 94: 181-195.

KAPPEN, L. 1964. Untersuchungen uber die Jahreslauf der Frost-, Hitze- und
Austrocknungsresistenz vin sporophyten einheimischer Polypodiaceen. Flora
155: 123-166.

KIELLAND, K. 1994. Amino acid absorption by artic plants: implications for plant
nutrition and nitrogen cycling. Ecology 75: 2373-2383.

1997. Role of free amino acids in the nitrogen economy of artic cryptogams.
Ecoscience 4: 75-79.

KLEKOWSKI, E. J. J. 1969. Reproductive biology of the pteridophyta. II. Theoretical
considerations. Botanical Journal of the Linnean Society 63: 347-359.

KLOCKARS, A. J., and G. SAX. 1986. Multiple comparisons. Sage Publications., Thousand
Oaks.

KNOWLES, R., and T. H. BLACKBURN. 1993. Nitrogen isotope techniques. Academic
Press, New York.

KOLB, K. J., and R. D. EVANS. 2003. Influence of nitrogen source and concentration on
nitrogen isotopic discrimination in two barley genotypes (Hordeum vulgare L.).
Plant Cell And Environment 26: 143 1-1440.

LIPSON, D., and T. NASHOLM. 2001. The unexpected versatility of plants: organic
nitrogen use and availability in terrestrial ecosystems. Oecologia 128: 305-316.

MILLER, J. H. 1968. Fern gametophytes as experimental material. Botanical Review 34:
361-&.

MOTTIER, D. M. 1927. The behavior of certain fern prothallia under prolonged
cultivation. Botanical Gazette 83: 244-266.

MULKEY, S. S., and R. W. PEARCY. 1992. Interactions between acclimation and
photoinhibition of photosynthesis of a tropical forest understory herb, Alocasia
macrorrhiza, during simulated canopy gap formation. Funct. Ecol. 6: 719-729.
NADELHOFFER, K. F., AND B. FRY. 1988. Controls On Natural N-15 And C-13
Abundances In Forest Soil Organic-Matter. Soil Science Society Of America
Journal 52: 1633-1640.










NADELHOFFER, K. J., and B. FRY. 1994. Nitrogen isotope studies in forest ecosystems. In
K. Lajtha and R. H. Michener [eds.], Stable Isotopes in ecology and
environmental science, 22-44. Blackwell scientific publications, Oxford.

NADKARNI, N. M., and T. MATELSON. 1991. Litter dynamics within the canopy of a
neotropical cloud forest, Monteverde, Costa Rica. Ecology 72: 849-860.

NAYAR, B. K., and S. KAUR. 1971. Gametophytes of homosporous ferns. The Botanical
Review 37: 295-396.

NIKLAS, K. 1997. The Evolutionary Biology of Plants. University of Chicago, Chicage.

NOBEL, P. S. 1978. Microhabitat, Water Relations, and Photosynthesis of a Desert Fern,
Notholaena-Parryi. Oecologia 31: 293-309.

OLIVER, M. J., B. D. MISHLER, and J. E. QUISENBERRY. 1993. Comparative Measures of
Desiccation-Tolerance in the Tortula-Ruralis Complex. 1. Variation in Damage
Control and Repair. American Journal ofBotany 80: 127-136.

OLIVER, M. J., Z. TUBA, and B. D. MISHLER. 2000. The evolution of vegetative
desiccation tolerance in land plants. Plant Ecology 151: 85-100.

ONG, B. L., and M. L. NG. 1998. Regeneration of drought-stressed gametophytes of the
epiphytic fern, Pyrrosia pilosellodes (L.) Price. Plant cell reports 18: 225-228.

PECK, J. H. 1980. Life history and reproductive biology of the ferns of Woodman Hollow,
Webster County, Iowa. Ph.D. Dissertation, Iowa State University, Ames.

PECK, J. H., C. J. PECK, and D. R. FARRAR. 1990. Influences of life history attributes on
formation of local and distant fern populations. American Fern Journal 80: 126-
142.

PICKETT, F. L. 1913. Resistance of the prothallia of Camptosorus rhizophyllus to
desiccation. Bul. Torrey Bot. Club 40: 641-645.

1914. Some ecological adaptations of certain fern prothallia Camptosorus
rhizophyllus Link., Asplenium platyneuron Oakes. American Journal ofBotany 1:
477-498.

1931. Notes on xerophytic ferns. America Fern Journal 21: 49-57.

PILLAI, R. S., and B. L. ONG. 1999. Effects of inorganic nitrogen availability on the
sporophytes of Acrostichum aureum L. Ph1,niy,,1nthetiiI 36: 259-266.









POREMBSKI, S., and W. BARTHLOTT. 2000. Granitic and gneissic outcrops (inselbergs) as
centers of diversity for desiccation-tolerant vascular plants. Plant Ecology 151:
19-28.

PRANGE, R. K., and D. P. ORMROD. 1982. Effects Of Ammonium And Nitrate Nutrition
On The Ostrich Fern (Matteuccia-Struthiopteris). Canadian Journal OfPlant
Science 62: 195-201.

PROCTOR, M. 2001. Patterns of desiccation tolerance and recovery in bryophytes. Plant
G1 ,fi I i Regulation 35: 147-156.

PROCTOR, M. C. F. 2000. The bryophyte paradox: tolerance of desiccation, evasion of
drought. Plant Ecology 151: 41-49.

2003. Comparative ecophysiological measurements on the light responses, water
relations and desiccation tolerance of the filmy ferns Hymenophyllum wilsonii
Hook. and H-tunbrigense (L.) Smith. Annals ofBotany 91: 717-727.

PROCTOR, M. C. F., and V. PENCE. 2002. Vegetative tissues: Bryophytes, vascular
resurrection plants and negative propagules. In M. Black and H. W. Prichard
[eds.], Desiccation and survival in plants: drying without dying, 207-237. CAB
International, Wallingford, UK.

PRYER, K. M., A. R. SMITH, and J. E. SKOG. 1995. Phylogenetic relationships of extant
ferns based on evidence from morphology and rbcL sequences. American Fern
Journal 85: 205-282.

PUTZ, F. E., and N. M. HOLBROOK. 1989. Strangler fig rooting habits and nutrient
relations in the llanos of Venezuela. American Journal ofBotany 76: 781-788.

RACUSEN, R. H. 2002. Early development in fern gametophytes: Interpreting the
transition to prothallial architecture in terms of coordinated photosynthate
production and osmotic ion uptake. Annals ofBotany 89: 227-240.

REICH, A., J. J. EWEL, N. M. NADKARNI, T. DAWSON, and R. D. EVANS. 2003. Nitrogen
isotope ratios shift with plant size in tropical bromeliads. Oecologia 137: 587-
590.

ROBINSON, D. 2001. delta N-15 as an integrator of the nitrogen cycle. Trends In Ecology
& Evolution 16: 153-162.

ROBINSON, S. A., J. WASLEY, M. POPP, and C. E. LOVELOCK. 2000. Desiccation tolerance
of three moss species from continental Antarctica. Australian Journal of Plant
Physiology 27: 379-388.









SANFORD, R. L., P. PAABY, J. C. LUVALL, and E. PHILLIPS. 1994. Climate,
geomorphology, and aquatic systems. In L. A. McDade, K. S. Bawa, H. A.
Hespenheide, and G. S. Hartshorn [eds.], La Selva: Ecology and Natural History
of a Neotropical Rain Forest, 19-33. University of Chicago Press, Chicago.

SAS-INSTITUTE. 2005. JMP User's Guide. SAS Instite, Inc., Cary.

SATO, T., and A. SAKAI. 1980. Freezing resistance of gametophytes of the temperate fern,
Polystichum retroso-paleaceum. Canadian journal of botany 58: 1144-1148 ill.

1981. Cold tolerance of gametophytes of some cool temperare ferns native to
Hokkaido. Can. J. Bot. 59: 604-608.

SCHEINER, S. M., and J. GUREVITCH. 2001. Design and analysis of ecological
experiments. Oxford University Press, Oxford.

SCHMIDT, A., and G. R. STEWART. 2003. Delta 15N values of tropical savanna and
monsoon forest species reflect root specilizations and soil nitrogen status.
Oecolgia 134: 569-577.

SCHMIDT, G., S. STUNTZ, and G. ZOTZ. 2001. Plant size: an ignored parameter in epiphyte
ecophysiology? Plant Ecology 153: 65-72.

SCHMIDT, S., and G. R. STEWART. 1999. Glycine metabolism by plant roots and its
occurrence in Australian plant communities. Australian Journal OfPlant
Physiology 26: 253-264.

SMITH, D. L. 1972a. Localization of phosphatases in young gametophytes of Polypodium
vulgare L. Protoplasma 74: 133-148.

1972b. Staining and osmotic properties of young gametophytes of Polypodium
vulgare L. and their bearing on rhizoid function. Protoplasma 74: 465-479.

TRESEDER, K. K., and P. M. VITOUSEK. 2001. Effects of soil nutrient availability on
investment in acquisition of N and P in Hawaiian rain forests. Ecology 82: 946-
954.

TUOMISTO, H., and K. RUOKOLAINEN. 1994a. Distribution of Pteridophyta and
Melastomataceae Along an Edaphic Gradient in an Amazonian Rain Forest.
Journal of Vegetation Science 5: 25-34.
1994b. Distribution of Pteridophyta and Melastomataceae Along an Edaphic
Gradient in an Amazonian Rain-Forest. Journal of Vegetation Science 5: 25-34.

TUOMISTO, H., and A. D. POULSEN. 1996. Influence of edaphic specialization on
pteridophyte distribution in neotropical rain forests. Journal of biogeography 23:
283-293.










TUOMISTO, H., and A. DALBERG. 1996. Influence of edaphic specialization on
pteridophyte distributions in neotropical rain forests. Journal ofBiogeography 23:
283-293.

TUOMISTO, H., and A. D. POULSEN. 2000. Pteridophyte diversity and species composition
in four Amazonian rain forests. Journal of vegetation science: official organ of
the International Association for Vegetation Science 11: 3 83-396.

TUOMISTO, H., A. D. POULSEN, and R. C. MORAN. 1998. Edaphic distribution of some
species of the fern genus Adiantum in western Amazonia. Biotropica 30: 392-
399.

WALKER, L. R., and G. H. APLET. 1994. Growth And Fertilization Responses Of
Hawaiian Tree Ferns. Biotropica 26: 378-383.

WALP, R. L. 1951. Fern Prothallia under Cultivation for 12 Years. Science 113: 128-129.

WANEK, W., S. K. ARNDT, W. HUBER, and M. POPP. 2002. Nitrogen nutrition during
ontogeny of hemiepiphytic Clusia species. Functional Plant Biology 29: 733-740.

WATANABE, M., T. KIKAWADA, N. MINAGAWA, F. YUKUHIRO, and T. OKUDA. 2002.
Mechanism allowing an insect to survive complete dehydration and extreme
temperatures. Journal OfExperimental Biology 205: 2799-2802.

WATKINS, J. E., and D. R. FARRAR. 2005. Origin and taxonomic affinities of Thelypteris
(subgen. Stegnogramma) burksiorum (Thelypteridaceae). Brittonia 57: 183-201.

WATKINS, J. E., C. CARDELUS, R. K. COLWELL, AND R. C. MORAN. 2006. Species
richness and distribution of ferns along an elevational gradient in Costa Rica.
American Journal ofBotany 93: 73-83.

WATKINS JR., J. E., and D. R. FARRAR. 2005. Origin and taxonomic affinities Thelypteris
(subgen. Stegnogramma) burksiorum (Thelypteridaceae). Brittonia 57: 183-201.

ZOTZ, G. 1997. Photosynthetic capacity increases with plant size. Botanica Acta 110:
306-308.
ZOTZ, G., A. ENSLIN, W. HARTUNG, and H. ZIEGLER. 2004. Physiological and anatomical
changes during the early ontogeny of the heteroblastic bromeliad, Vriesea
sanguinolenta, do not concur with the morphological change from atmospheric to
tank form. Plant Cell And Environment 27: 1341-1350.















BIOGRAPHICAL SKETCH

James Edward (Eddie) Watkins, Jr. was born on 12 March, 1974 in Ozark, Dale

County, Alabama. He attended the now-condemned Flowers Elementary School where

his first introduction to science was a project on turtles by his first grade teacher Mrs.

Hopper. He graduated to attend D.A. Smith Middle School. It was there, under direction

of his eighth grade teacher, Dena Byers he first began to develop an understanding of

scientific experiment. He competed in several regional science fairs, making it as far at

the Alabama State Science Fair for his work on factors controlling the rate at which mice

could exit a complicated maze.

During these years, he spent much of his time fishing, hunting, building forts, and

observing nature from his daily hikes in to the forests surrounding his home. During these

early years he developed a true connection with the natural world. He gained his early

understanding of how this world was put together by his first biological mentor Ms.

Linda Dees: science teacher at Carroll High School. During these formative high school

years he began to put together what he would do for the rest of his life. One of the most

important developments came when Ms. Dees required his freshman biology class to

complete a plant collection. This Eddie did by only collecting live ferns that were then

transplanted into the school's nature preserve. In the course of this collection, he

discovered two of the rarest ferns in Alabama and went on to publish some of this work

in a peer-reviewed journal. During his early fern forays, he made the acquaintance of

Professor Warren Herb Wagner, Jr.; the world's leading fern authority at the time, and









continues to inspire Eddie's work today. After graduation, he attended Auburn

University, where he worked under the direction of Dr. John D. Freeman and Dr. Robert

S. Boyd. In the lab and courses of Dr. Boyd that Eddie was finally able to put nature and

experimental sciences together and begin to develop an understanding of the complexities

of ecology. After graduation, he attended Iowa State University under Dr. Donald Farrar

where he attained an MS degree with his thesis on Thelypteris burksiorum. These years

were paramount to his Pteridological development, as Dr. Farrar is one of the last old-

school fern biologists and was as smitten with ferns as Eddie. Eddie graduated in 2000

and spent a year living with his wife Catherine in Costa Rica, studying the magnificent

array of ferns at La Selva Biological Station and beyond. He then returned to the South

where he began his doctoral studies with Dr. Stephen Mulkey and Dr. Michelle C. Mack.

After completing of his doctoral studies, Eddie will begin a post doctoral fellowship in

the lab of Michelle Holbrook at Harvard University.