ecological factors affecting the diversity of …

i

ECOLOGICAL FACTORS AFFECTING THE DIVERSITY OF TROPICAL TREE SEEDLINGS

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and

Agricultural and Mechanical Collegein partial fulfillment of the

requirements for the degree ofDoctor of Philosophy

in

The Department of Biological Sciences

byC. E. Timothy Paine

B. A. Dartmouth College 1999August, 2007

ii

DEDICATION

In memory of my grandfatherHugh R. (Farf) Gibson

1920 – 1999

iii

ACKNOWLEDGEMENTS

I first want to thank Kyle Harms. My advisor has been consistently kind, considerate, and

supportive of my goals. In his gentle way, he has shepherded me away from dozens of unpro-

ductive dark alleys, and encouraged me to head out independently toward the edge of scientific

knowledge. He is one of a kind.

I thank my graduate committee. Discussions with Richard Stevens, Jim Cronin, Jay

Geaghan, and Thomas Dean sharpened the arguments forwarded in this dissertation. Their guid-

ance has been invaluable.

Luis Escobar, James P. Geaghan and Qingfang Wu were generous with their time and

knowledge, and helped develop the data analysis. Extensive discussions with Barry Moser, who

died unexpectedly in 2006, brought much of statistics into focus for me.

I deeply appreciate the efforts of Pamela Weisenhorn, Armando Mendoza, Christina

Georghiou, Nataly Hidalgo, Kristen Schmitt, and Cecilia Carrasco, who worked with me as field

assistants. All tolerated many hot hours in the muddy, buggy understory. Pamela Weisenhorn, par-

ticularly, far surpassed what could be expected from a field assistant. Her eye for detail improved

the design and execution of these experiments.

I want to thank my friends and labmates at LSU, who have been supportive and wonder-

ful, including Heather Passmore, Paul Gagnon, Jonathan Myers, and Adriana Bravo. I owe special

thanks to Jane Carlson, who started at LSU with me, and has been my best friend and confidante

since the first. Jane Carlson, Jonathan Myers, Heather Jackson, Margaret Metz, Sarahfaye Mahon,

and Heather Jackson carefully read drafts of the chapters of this dissertation, and made many sug-

gestions that improved it. The errors that remain are my own.

iv

I appreciate my colleagues and friends from Cocha Cashu. Susan Willson took a chance

on me, and introduced me to Cocha Cashu. John Bunce, Varun Swamy, Kyle Dexter, Natalia

Quinteros, and Nicole Gibson supported me, and each other, throughout our concurrent Ph.D.

investigations in el bosque encantado. Mercedes Foster provided guidance and judgement, and

un monton of natural history. Working with seedlings would have been impossible without the

plant identification skills shared by her, John Terborgh, Nigel Pitman and Fernando Cornejo. Mac

Alford (USM) determined the Casearia sp. nov. used in Chapter 4. I wish to single out Patricia

Alvarez for special thanks – her energy, drive, and laughter made life in the jungle exciting and

joyous, and made the manuplants.org project possible. I continue to miss Francis Bossuyt and

Abbie Sorenson, whose love of all natural things opened my eyes.

I thank my family of Paines, Feldermans, Fauses, Gibsons, and others. Their love was

unconditional, whether or not they knew what I spent all those years doing in the jungle. More

than anything, I thank Sarahfaye Mahon for her incredible support, encouragement, and compan-

ionship.

Collaboration with Stefan Schnitzer and Walt Carson made possible the experiments

described in Chapter 3. Christopher Baraloto and Margaret Metz also kindly shared unpublished

results that contributed to the generality of Chapter 3. And collaborating with Harald Beck, who

has long been a role model for me, made the mammal-exclosure experiments of Chapter 4 possi-

ble. Discussions with him also materially improved Chapter 3. John Terborgh has been extremely

generous in sharing unpublished results from his seed-trapping project. These data provided con-

text that strengthened all three experiments of this dissertation.

I thank Peru’s Instituto Nacional de Recursos Naturales (INRENA) for permission to con-

duct research in Manu National Park. This research was supported through a LASPAU fellowship

v

from the Organization of American States, a Grant-in-Aid-of-Research from Sigma Xi, and the

LSU Biology Graduate Student Association (BioGrads). I particularly appreciate the support of a

LSU Board of Regents fellowship, without which this dissertation would not have been possible.

vi

TABLE OF CONTENTS

DEDICATION ................................................................................................................................ ii

ACKNOWLEDGEMENTS ........................................................................................................... iii

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

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

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

CHAPTER 1 – INTRODUCTION ..................................................................................................1PROCESSES THAT LIMIT THE RECRUITMENT OF TROPICAL TREE

SEEDLINGS ........................................................................................................................1LIMITS TO DISPERSAL ........................................................................................2POST-DISPERSAL LIMITS TO ESTABLISHMENT ............................................5SITE DESCRIPTION ............................................................................................10FOCAL SPECIES ..................................................................................................16DISSERTATION STRUCTURE ...........................................................................16

CHAPTER 2 — WHAT SHAPES TROPICAL SEEDLING COMMUNITY STRUCTURE? SEED DISPERSAL VERSUS ENVIRONMENTAL CONDITIONS ..............................21

INTRODUCTION .................................................................................................21METHODS ............................................................................................................23RESULTS ...............................................................................................................31DISCUSSION ........................................................................................................40

CHAPTER 3 — WEAK COMPETITION AMONG TROPICAL TREE SEEDLINGS: IMPLICATIONS FOR SPECIES COEXISTENCE ..........................................................47

INTRODUCTION .................................................................................................47MATERIALS AND METHODS ..........................................................................49RESULTS ...............................................................................................................55DISCUSSION ........................................................................................................59

CHAPTER 4 — SEED PREDATION BY NEOTROPICAL RAINFOREST MAMMALS INCREASES DIVERSITY IN SEEDLING RECRUITMENT .........................................64

INTRODUCTION .................................................................................................64METHODS ............................................................................................................66RESULTS ...............................................................................................................74DISCUSSION ........................................................................................................81CONCLUSIONS ...................................................................................................85

vii

CHAPTER 5 – CONCLUSIONS ..................................................................................................86SUMMARY ...........................................................................................................86SYNTHESIS ..........................................................................................................86

LITERATURE CITED ...................................................................................................................89

APPENDIX 1 – DENSITIES OF SEEDLINGS IN SEED-ADDITION EXPERIMENTS OF CHAPTER 2 ....................................................................................................................107

APPENDIX 2 – LETTERS OF PERMISSION ...........................................................................118

VITA ..........................................................................................................................................122

viii

LIST OF TABLES

Table 1.1. Characteristics of the 18 species used in this dissertation ........................................... 17

Table 2.1. Species sown in the seed-addition project, sorted by family ...................................... 25

Table 2.2. Dispersal, not establishment, predicts seedling species composition, even 2+ years after a one-time addition of seeds ..................................................................................... 39

Table 3.1. Seedling densities in high-density experimental plots and in 11 Neotropical rainforests ......................................................................................................................... 49

Table 4.1. Species sown in mammal exclosures, sorted by family .............................................. 69

Table 4.2. Overall treatment effects of mammal exclosures, seed species, and their interactions ....................................................................................................................... 75

ix

LIST OF FIGURES

Figure 1.1. Conceptual diagram of the fates of tree seedlings that constitute a cohort ................. 2

Figure 1.2. Map of Estación Biológica Cocha Cashu, Manu National Park, Peru (EBCC), illustrating the sites of the experiments that constitute this dissertation ........................... 12

Figure 1.3. Seasonal trends in rainfall (bars) and maximum and minimum temperature (lines) at EBCC, displayed as a mean rain-year ...............................................................................15

Figure 2.1. Enviromental similarity was weakly and inversely, but significantly, correlated with geographic distance ........................................................................................................... 30

Figure 2.2. (a) Stem density and (b) species richness of focal and non-focal seedlings over three years .................................................................................................................................. 32

Figure 2.3. Density of recruited seedlings of eight species over time, by sown density ............. 34

Figure 2.4. The variance in focal species seedling stem density was better explained by seed dispersal than by environmental conditions ......................................................................35

Figure 2.5. Species richness of focal species, the cohort of 2004, and all seedlings together. Each line represents ....................................................................................................................37

Figure 2.6. Variance in species richness of focal seedlings, the cohort of 2004, and the overall seedling layer was explained by seed dispersal, environmental conditions, and their interactions ........................................................................................................................ 38

Figure 3.1. Relative height growth rate (RGRht) of (a) Brosimum alicastrum, (b) Pouteria reticulata and (c) Matisia cordata seedlings grown in plots of experimentally reduced density ................................................................................................................................56

Figure 3.2. Survival of (a) Brosimum alicastrum, (b) Pouteria reticulata and (c) Matisia cordata seedlings grown in plots of experimentally reduced density .............................................57

Figure 3.3. Histograms of above ground zone of influence (ZOI) overlaps in (A) high-density, Matisia-dominated plots and (B) typical mixed-species plots .......................................... 58

Figure 4.1. Schematic of the five different types of mammal exclosures .................................... 70

Figure 4.2. The survival of seeds and resultant seedlings differed widely among the five types of mammal exclosures ...........................................................................................................75

x

Figure 4.3. Aspects of seedling recruitment in exclosures open or closed to each group of mammals ............................................................................................................................77

Figure 4.4. Interacting effects of adult abundance and small- and medium-mammal access on seed survival and seedling recruitment ............................................................................. 79

Figure 4.5. Interacting effects of seed mass and small- and medium- mammal access on seed survival and seedling recruitment ..................................................................................... 80

xi

ABSTRACT

Seed dispersal and seedling establishment – the two stages in seedling recruitment – set

the spatiotemporal distribution of new individuals in plant communities. Diversity often increases

at the seed to seedling transition, making it critical for species coexistence. Debate continues

regarding the effects of each stage on the community structure of diverse forests. Neutral theories

postulate a strong role of dispersal, whereas niche-differentiation theories suggest that envi-

ronmental conditions may be more important. This dissertation tested the effects of dispersal,

competition and predation on the structure of the seedling layer in a pristine Amazonian rainfor-

est.

Seed-addition experiments broadly tested the relative importance of dispersal and envi-

ronmental conditions on seedling community structure. Dispersal treatments explained more

variance in community structure than did environmental conditions. This was the first vari-

ance-partitioning study to show that dispersal affects not only seedling density, but also diversity

and species composition. Two more narrowly focused studies tested the intensity of competition

among seedlings, and examined the effects of various mammalian predators on seedling recruit-

ment. Evidence for inter-seedling competition was weak: individual growth and survival rates

were generally unrelated to stem density, and seedlings’ zones of influence rarely overlapped

substantially. As predators, small and medium-sized mammals reduced seedling density, whereas

large mammals had no detectable effects. Furthermore, small mammals generated a rare-species

advantage, the fundamental element of frequency dependence.

Integrating the three studies, we suggest that dispersal is more important for seedling

community structure than are environmental conditions. Given the low density of seedlings in

xii

Neotropical forests, we infer that competition among tree seedlings is largely irrelevant to their

recruitment. Seed predators operated in a distinctly non-neutral manner, preferentially removing

seeds of common and large-seeded species. Despite the powerful effects of predation, dispersal

explained more variance in seedling recruitment than did all aspects of environmental variation

(including predation). Taken together, the results of these three experiments support a view that, at

least for young plants, and at small scales, dispersal may more strongly influence the species com-

position of tropical trees than environmental conditions, consistent with predictions from neutral

models.

1

CHAPTER 1 – INTRODUCTION

The objective of this dissertation is to examine the effects of three ecological processes

that shape plant community structure: seed dispersal, inter-seedling competition, and mammalian

predation. This introductory chapter provides context to bind together the experiments used to

examine each process, and a framework within which their results may be interpreted. In the

first section, I sequentially review the stages of the recruitment of seedlings from seeds, with an

emphasis on those processes that limit recruitment. Much of this section was previously published

as an invited review paper in Ecosistemas, a Spanish-language journal of ecology (Harms and

Paine 2003). In the second section, I provide a brief overview of the physical and geographic set-

ting of Estación Biológica Cocha Cashu, where all research was conducted. I then introduce the

eighteen species that were used as seeds and seedlings in the experiments. I conclude by summa-

rizing the objectives of each of the dissertation’s chapters.

PROCESSES THAT LIMIT THE RECRUITMENT OF TROPICAL TREE SEEDLINGS*

Seedling recruitment consists of two stages: the dispersal of seeds from adult plants, and

the subsequent germination and survival of seedlings. The stages of seedling recruitment are pre-

sented diagrammatically in Figure 1.1. Recruitment can be limited at any of a plant’s sequential

life-cycle stages (reviewed in Muller-Landau et al. 2002, Munzbergova and Herben 2005). Prior

to dispersal, a scarcity of adults, low adult fecundity, or pre-dispersal seed predators may limit the

number of seeds available for dispersal. At the dispersal stage, the failure of seeds to arrive to a

* This section reprinted by permission from Ecosistemas. It has been translated from Spanish to English, and lightly edited for content. Citation: Harms, K. E. y Paine, C. E. T. 2003. Regeneración de árboles tropicales e implicaciones para el manejo de bosques naturales. Ecosistemas 2003/3.

2

given site may limit recruitment. Dispersal limitation, the consequence of these three processes,

regulates the initial spatiotemporal distribution of a cohort of seeds and may have substantial

consequences for community structure and dynamics (Levine and Murrell 2003). Post dispersal,

abiotic limitations and biotic interactions eliminate individuals from the community, a process

referred to as establishment limitation or ecological filtration (Harper 1977, Muller-Landau et al.

2002).

LIMITS TO DISPERSAL

The arrival of seeds to a certain site may be limited by adult dispersion and fecundity,

pre-dispersal seed predation, or vagaries of the seed dispersal vector. Local rarity and clumping

at most spatial scales are common in forests worldwide (He et al. 1997, Pitman et al. 1999, Condit

et al. 2000, Plotkin et al. 2002). Pre-dispersal seed predators (as well as frugivores that damage

��

��

��

��

��

Figure 1.1. Conceptual diagram of the fates of tree seedlings that constitute a cohort. In general, the intensity of ecological filtration decreases as the individuals of a cohort grow and age. Many processes contribute to ecological filtration, but I highlight competition and predation, as they are the foci of Chapters 3 and 4, respectively.

3

seeds) are capable of reducing seed crops (Hulme 2001). In seed-trapping studies, the large seeds

often bear evidence of pre-dispersal attack by seed-feeding insects (e.g., Forget et al. 1999). Adult

dispersion patterns and pre-dispersal seed predation together impose source limitation (sensu

Muller-Landau et al. 2002) on seedling recruitment by limiting the number of viable seeds avail-

able for dispersal.

Birds, mammals, wind, water, or other vectors may disperse seeds from adult trees (Levin

et al. 2003). The essential concept of dispersal limitation is that these vectors do not transport

seeds to all possible sites for germination. Dispersal limitation – the failure of seeds to arrive to

all possible establishment sites – is frequently severe. In a ten-year seed-trapping study on Barro

Colorado Island, 50 of 260 tree species in the community failed to disperse even a single seed to

any one of 200 seed traps (Hubbell et al. 1999). Investigating pioneer species in Panama, Dalling

et al. (2002) found that all but the smallest-seeded, most fecund species were highly dispersal-

limited over four years of study. An extreme case of limited dispersal occurs in the Brazil nut,

Bertholletia excelsa (Lecythidaceae), in which indehiscent woody fruit fall by gravity and are

gnawed open by agoutis (Dasyprocta spp.). Agoutis cache the seeds, but on average, disperse

them less than 5 m (Peres and Baider 1997). In general, seed dispersal is often distance-restricted,

in that most seeds are dispersed to sites relatively near adults. At scales that include the seed shad-

ows of multiple trees, seed dispersal generates clumped patterns of seed rain that are centered on

adults (Clark et al. 1998, Clark et al. 1999, Schupp et al. 2002).

Though the density of seed rain declines monotonically with distance from parent trees,

depositional biases in dispersal processes may also create smaller-scale clumping in seed rain.

For example, the movement patterns of vertebrate frugivores affect the dispersal of seeds because

frugivores are likely to defecate where they spend time feeding, i.e., at fruiting trees (Schupp et

4

al. 2002). During the course of my Ph.D. graduate research, I co-advised a Swiss masters student

who tested this hypothesis, and found that frugivore aggregations in fruiting trees generate hot-

spots in seed rain, in terms of both the density and the species richness of dispersed seeds (Baiker

2007). The primary dispersers of Virola calophylla (Myristicaceae) seeds are spider monkeys

(Ateles bezelbuth, Russo 2005). Most Virola seeds fall beneath maternal adults, but vertebrate-

dispersed seeds are deposited disproportionately beneath Ateles sleeping sites, as compared to

in-transit rest sites or random locations in the understory (Russo and Augspurger 2004). Male

bellbirds (Procnias tricarunculata) in Costa Rican montane forests typically perch and defecate

at gap edges, a preferential site for seed germination (Wenny and Levey 1998). Even the wind

may deposit seeds contagiously. Local updrafts, eddies, and other fluid-dynamic features of the

winds in and above forests deposit seeds in clumped patterns (Horn et al. 2001, Nathan et al.

2002). Clumped patterns in seed dispersal may persist through ontogeny, affecting the distribu-

tion of adult trees (Condit et al. 2000, Seidler and Plotkin 2006). Seed dispersal is a stochastic

process that generates distance-limited distributions of seeds that are clumped at various spatial

scales.

Evidence from simulations and human-affected forests indicate the importance of an

intact fauna to ensure effective seed dispersal and recruitment of seedlings. A simulation param-

eterized by extensive field data from Borneo suggested that a loss of frugivorous vertebrates

would curtail seed dispersal and reduce seedling diversity by 60% in the stand as a whole (Webb

and Peart 2001). Comparing hunted and pristine forests in Amazonian Peru, Terborgh et al. (in

review) found that hunting, particularly for seed-dispersing monkeys, lowered sapling stem den-

sity overall, and increased the relative abundance of species dispersed by abiotic vectors. At a

5

broader scale, countless successful species introductions to sites outside their native range indicate

that dispersal can limit species distributions.

These studies indicate that dispersal may limit recruitment by setting the initial spatiotem-

poral pattern of seedlings. Nevertheless, this limitation may be of little importance if variation

in environmental conditions strongly affects seedling growth and survival (Nathan and Muller-

Landau 2000, Zobel et al. 2000). I weigh the relative effects of seed dispersal and post-dispersal

environmental conditions on the density, diversity, and species composition of the seedling layer

in Chapter 2.

POST-DISPERSAL LIMITS TO ESTABLISHMENT

Once seeds are dispersed to a given site, their establishment is not assured (Schupp et al.

2002). Potential limits to seedling establishment fall into three categories: abiotic factors, biotic

interactions, and intrinsic characteristics of seeds themselves (Nathan and Muller-Landau 2000,

Muller-Landau et al. 2002). Important abiotic factors include the availability of light, water, and

soil nutrients. Important biotic factors are the intensity of competition, both from nearby canopy

adults and neighboring seedlings, as well as the density and host specificity of natural enemies,

such as post-dispersal seed predators, pathogens (especially fungi), and herbivores. The absence of

appropriate mutualists, such as mycorrhizal fungi, may also limit establishment in some habitats

(Klironomos 2002, Moora et al. 2004). These post-dispersal limits to establishment are discussed

in detail in the following sections.

Abiotic Factors

Abiotic factors, such as the availability of light, water and soil nutrients may influence the

germination of seeds and survival of seedlings. Of these three, light is probably the most impor-

tant. Decreased mortality in high-light sites was demonstrated by transplantation of seedlings into

6

a range of light environments in Costa Rica (Kobe 1999). However, the abundance of seedlings

was only weakly related to light availability in another study from the same year and region

(Nicotra et al. 1999). Canopy gaps provide high-light, high-temperature conditions, and they were

at one time believed to play a major role in maintaining tropical forest tree diversity (Denslow

1980), though today the importance of that role is contended (Denslow et al. 1990, Hubbell et al.

1999, Schnitzer and Carson 2001). Nevertheless some (and perhaps most) tree species require the

high-light environment of gaps to recruit saplings into the canopy (Hartshorn 1980, King 1994).

Soil nutrients may affect the growth, and indirectly the mortality, of rainforest trees

(Grubb 1977). In a recent meta-analysis, fertilization increased growth in a majority of species,

with pioneer species responding more strongly than shade-tolerant species (Lawrence 2003).

In a common-garden experiment in Ghana, seedlings grown in soil from wet evergreen forest

had lower foliar concentrations of N, P, K Ca and Mg than those grown in soil from a moist

semi-deciduous forest, reflecting differences in available soil nutrients (Veenendaal et al. 1996).

Nevertheless, the effects of fertilization on seedlings can be difficult to predict in detail (Burslem

et al. 1995), probably because the effects are frequently indirect, mediated by nutrient cycling

through detritivore food webs (Lawrence 2003).

The availability of water varies seasonally in many tropical forests. Seedlings may be

susceptible to seasonal drought, because their roots typically penetrate only shallowly into soil,

whereas larger plants may access deeper sources of water (Engelbrecht et al. 2005). When water

is scarce, it may be preemptively consumed by neighboring adults (Casper and Jackson 1997,

Coomes and Grubb 2000). Rainforest tree seedlings differ in their tolerance of drought condi-

tions (Engelbrecht and Kursar 2003), in part because of differences in allometry and architecture

(Yamada et al. 2005). The effects of water availability on individual performance may scale up to

7

affect the diversity of tropical forests. Differences in seasonality or drought intensity at regional

scales may generate large-scale patterns in diversity. Alwyn Gentry, in assembling his massive

dataset on the composition of tropical forests, noted that species richness increases with annual

rainfall, up to an asymptote of almost 300 species per hectare at 6 meters of rain per year (Gentry

1988). I did not experimentally test abiotic influences on seedling recruitment in this dissertation.

But as an additional project, I manipulated water availability in a seasonal tropical forest and

observed the effects on seedling recruitment (Paine, Harms and Ramos, in prep).

Biotic Interactions

There is abundant evidence that adult trees impose above- and below-ground resource

competition for the abiotic resources discussed above, which suppresses seedling growth and

survival (Coomes and Grubb 2000, Lewis and Tanner 2000, Barbaris and Tanner 2005). Indeed,

the essence of the gap paradigm, that canopy gaps are essential to the successful recruitment

of tree seedlings, is predicated on intense asymmetrical competition imposed by adults on

seedlings (Denslow 1980, Ostertag 1998). The degree to which seedlings additionally compete

among themselves is poorly understood, but could strongly affect their recruitment. If the most

intense competition comes from nearby adults, then the identity of neighboring seedlings would

be relatively unimportant to individual seedling growth and survival. On the other hand, if

seedlings of relatively similar sizes compete intensely among themselves, then the identity of

neighboring seedlings may be an additional, critical determinant of seedling performance. In this

case, seedlings could partition resources among themselves (Tilman 1982). I test the intensity of

competition among seedlings in Chapter 3, in a collaborative effort with investigators working on

Barro Colorado Island, Panama.

8

Seed predation also can influence seedling establishment at varying spatiotemporal

scales. In a mast-fruiting dipterocarp forest, a seed-sowing experiment showed that, except during

supra-annual mast fruiting events, seed predators consumed almost all seeds (Curran and Webb

2000). Satiation of the wild pigs (Sus scrofa), and therefore recruitment, occurred only in mast

years, when all dipterocarp species fruited synchronously (Curran and Leighton 2000). Islands

in Gatún Lake (part of the Panama Canal) support varied subsets of the native mammalian com-

munity. Seed survival was greatest on mammal-free islands, and similarly low in sites with intact

mammal communities or only rats (Asquith and Mejia-Chang 2005). This suggests that small

mammals may reduce seed survival to a degree disproportionate to their low biomass. Because

seed predators frequently aggregate on high-density resource patches, seed predation is often posi-

tively density-dependent and may increase diversity. For example, per-capita survival of seeds of

Virola calophylla dispersed in groups was inversely related to the number of seeds in each group

(Russo 2005). I experimentally test the effects of predation by three size-classes of terrestrial

mammals on the density and diversity of the seedling layer in Chapter 4.

Like seed predation, herbivory on seedlings is highly variable, but can be important in

shaping community structure (Huntly 1991). Plant species differ in their allocation of resources

to growth, reproduction and defense, leading to differing rates of herbivory among species (Coley

1983, Coley and Barone 1996, Howlett and Davidson 2001). Herbivory was the most important

source of mortality for newly-germinated seedlings in a tropical rainforest (Dalling and Hubbell

2002) and in a variety of other habitats (Moles and Westoby 2004b). Additionally, herbivores and

other terrestrial mammals may damage seedlings in dynamic tropical forests by trampling them

(Clark and Clark 1989, Roldan and Simonetti 2001). In addition to testing their effects as seed

predators, I also test the effects of terrestrial mammals as herbivores of seedlings in Chapter 4.

9

Many natural enemies are host specific at the species, genus, or family level (Coley and

Barone 1996, Novotny et al. 2002) and the intensity of their attack frequently decreases with

distance from conspecific adults (e.g. Blundell and Peart 1998). These observations, coupled with

distance-restricted seed dispersal, form the essence of the Janzen-Connell hypothesis of the main-

tenance of species diversity (Janzen 1970, Connell 1971). Tests of this hypothesis, which predicts

that seedlings recruit most successfully at some distance away from adult conspecifics, thus main-

taining high diversity, have had mixed results, but on the whole are supportive (Hammond and

Brown 1998, Hyatt et al. 2003, Terborgh et al. in review). The Janzen-Connell model is a special

case of negative density-dependence, in which demographic success is inversely related to popula-

tion size. Rare species thus have a demographic advantage over common species. Negative density

dependence is a diversifying influence in many diverse habitats (Webb and Peart 1999, Harms et

al. 2000, Hille Ris Lambers et al. 2002, Wills et al. 2006). I also test the degree to which mam-

malian predation is a source of density-dependent mortality for seeds and seedlings in Chapter 4.

Seed Characteristics

Seeds vary widely in their probability of establishment (Garwood 1983, Foster and Janson

1985). Larger seeds and those with hypogeal cotyledons have higher probabilities of establishment

than those with small seeds or photosynthetic cotyledons, particularly in the shaded understory

(Foster and Janson 1985, Silman 1996, Dalling and Hubbell 2002, Paz and Martinez-Ramos

2003). Large seeds can be more susceptible to seed predation (Moles et al. 2003), but are also

more likely to resprout and recover from physical disturbances (Harms and Dalling 1997). I also

test the interaction of mammalian predation and seed size in Chapter 4.

10

Interactions among the Stages of Seedling Recruitment

Though this review has summarized the stages of seedling recruitment sequentially, they

are not independent of each other. For example, the distribution of the seed rain has consequences

for post-dispersal seed predation. Small-scale clumps in the distribution of seeds, which can result

from the behavior of seed dispersers (Russo and Augspurger 2004), may be particularly attractive

to density-dependent herbivores or seed predators (Blundell and Peart 1998, Russo 2005). Thus

clumping in seed dispersal may be correlated with low rates of seedling establishment. Seed

traits are another source of correlations among the stages of seedling recruitment. For example,

large seed size may simultaneously limit the set of dispersal vectors to large-bodied vertebrates,

decrease mean dispersal distances, increase germination rates, and increase tolerance to seed

damage and seedling herbivory (Dalling and Harms 1999, Russo 2003, Moles and Westoby

2004a, b).

SITE DESCRIPTION

All research for this dissertation was conducted at Estación Biológica Cocha Cashu

(EBCC) in Peru’s 2-million-hectare Manu National Park, a UNSECO world heritage site (Figure

1.2). The park was established in 1973 by a National Supreme Decree of Peru’s legislature. Manu

is the cornerstone of a chain of parks and other protected areas that stretch south along the foot

of the Andes to Bolivia’s Madidi and Amboró National Parks, referred to as the Vilcabamba-

Amboró Forest Ecosystem Corridor. EBCC is unique among field research stations in that it is

both pristine and provides access to a veriety of incredibly diverse habitats. Sites untrammeled by

anthropogenic impacts are increasingly rare in tropical forests and around the globe. Research at

EBCC offers a spectacular window into patterns and processes that are obscured elsewhere by the

11

residue of man. The unspoiled diversity of EBCC has long been a magnet for investigators in ecol-

ogy and other sciences (Terborgh 1983, Gentry 1990).

Because of its isolation, and the protection afforded by the park, EBCC provides access

to one of the most pristine forests in the world. The site has been only minimally impacted by

anthropogenic activities. Only minimal subsistence hunting by Matsigenka and un-contacted

native peoples has occurred at EBCC in at least 35 years1. Though selective logging for mahogany

(Swietenia macrophylla, Meliaceae) occurred in the Manu watershed, individuals of that species

persist close to the station. Rubber trees (Hevea guianensis, Euphorbiaceae) are scarce in the

floodplain of the Manu, so the rubber boom of the 1940’s largely passed it by. The Manu water-

shed was mapped by the Cities Service Corporation (now CITGO) in 1972-73, but no petroleum

exploration or extraction took place near EBCC. No invasive species are known from the forest

near EBCC (though a few species are cultivated for food by station residents).

The diversity of the flora and fauna in the habitats surrounding EBCC is entrancing, and

at times overwhelming. The point (alpha) diversity of birds can exceed 160 species (Terborgh et

al. 1990). Thirteen species of monkeys and marmosets are known from EBCC, nine of which

easily can be seen (or heard) on a daily basis (Janson and Emmons 1990, C. E. T. Paine, personal

observation). Large mammalian predators, including jaguars, giant otters, and short-eared dogs

(Panthera onca, Pteroneura brasiliensis, and Atelocynus microtis, respectively), are frequently

observed. Extensive collecting, primarily by Dr. Robin Foster revealed 2,874 species of plants

from Manu National Park (Foster 1990). The forests of the low-lying, relatively recent floodplain

alone contain at least 1,372 species of plants (Foster 1990). A census of twenty-one permanent 1 Notably, a group of un-contacted natives camped approximately 3 km from EBCC between April and May 2005. Later visits to their campsite revealed bones of tapirs (Tapirus terrestris) and spider monkeys (Ateles bezelbuth), and shells of tortoises (Geochelone denticulata), giving evidence of their hunting.

12

Figure 1.2. Map of Estación Biológica Cocha Cashu, Manu National Park, Peru (EBCC), illustrating the sites of the experiments that constitute this dissertation. The open circle in the inset map indicates the location of Manu National Park in southeastern Peru. EBCC maintains about 60 km of trails (solid lines), which provide unparalleled access to pristine, diverse forest and other habitats. Base map produced by Nelson Gutiérrez, Amazonian Conservation Association.

13

Figure 1.2.

14

tree plots (~36 ha in total) enumerated a total of 825 species of canopy-attaining trees (Pitman

et al. 1999). A single hectare of floodplain forest contains 174 species of trees on average (range:

126 – 217 species, Pitman et al. 2001). Similar statistics from North America provide a striking

contrast. Half-hectare samples of southeastern Texas costal forest contained an average of 15 tree

species (range: 3-23 species, Marks and Harcombe 1981). A total of 253 tree species are found in

the moist temperate forests of eastern North America (Latham and Ricklefs 1993), and the entire

continental United States is estimated to contain 20,000 native vascular plant species (Davies et al.

1986).

Climate

Estación Biológica Cocha Cashu is located in a seasonally dry tropical rainforest. It is

classified as a Tropical Moist Forest in the scheme of Holdridge (1947). At the station, rainfall is

recorded manually with a simple cylindrical rain gauge located in the center of a ~0.5-ha clearing.

Including only those 15 years in which rainfall data were recorded in 9 or more months between

1984 (when record-keeping began at the site) and 2006, mean annual rainfall was 2167 mm

(range: 652 – 2763 mm; Figure 1.3). Annual calculations of rainfall may be misleading as the wet

season (October – April) is split between two years. A more biologically relevant rainfall estimate

may be generated with “rain-years” that run from May to April of the following calendar year.

Mean annual rainfall calculated on this basis (including only rain-years with more than 9 months

of observations) is 2339 mm. The driest rain-year on record occurred in 2004-05 (364 mm), and

the wettest (3352 mm) occurred in 1996-97. Intense wet-season rains can cause the Manu River

to flood. Floods typically last only 1-3 days, since the catchment area of the Manu watershed,

15

0

5

10

15

20

July October January April

Mea

n D

aily

Rai

nfa

ll (m

m)

0

5

10

15

20

25

30

Max

imum

and M

inim

um

Temperatu

res (ºC)

Figure 1.3. Seasonal trends in rainfall (bars) and maximum and minimum temperature (lines) at EBCC, displayed as a mean rain-year. Seasonal patterns in rainfall are much stronger than those in temperature.

upstream of EBCC, is relatively small. None of my study plots was inundated by floodwaters in

the course of this dissertation.

In contrast to rainfall, seasonal patterns in temperature are moderate. Diel temperature

fluctuations are on the same order of magnitude as seasonal variation. Daily minimal and maxi-

mal temperatures are recorded at the station using a U-tube type recording thermometer in the

forest understory, approximately 10 m from the station clearing. A rough mean daily temperature

may be calculated as the mean of these extreme values. This “mean” temperature ranged from

24.2 °C in January to 21.8 °C in July. The average diel temperature fluctuation is slight in January

(2.6 °C), and greater in July (4.5 °C).

16

Despite the seasonal stability of air temperatures at EBCC, friajes are a common element

of the dry-season weather conditions. These cold fronts sweep up from Antarctica and southern

South America and across southern Amazonia, bringing several days of cold weather, and often

rain. Several friajes, of varying intensity, occur each year between June and September. The cold-

est temperature on record (8.5 °C) at EBCC occurred in August 1993. The minimum temperature

of the previous day had been 9.5 °C. Minimum daily temperatures of < 10 °C have been recorded

only 4 times (out of 5168 observation days). Nevertheless, these rare events may have profound

effects on the survival and species composition of the flora and fauna, much of which lacks behav-

ioral or physiological defenses against cold (e.g., Olmsted et al. 1993).

FOCAL SPECIES

Eighteen species were used in the three experiments that constitute this dissertation,

representing 14 families (Table 1.1). I used 8 species to compare the effects of dispersal and estab-

lishment in Chapter 2. Seven of these, and another 7 species besides, were used in the predation

experiments of Chapter 4. Due to logistical requirements, the three species used to study intra-

seedling competition in Chapter 3 were distinct from those studied in other experiments. Criteria

for inclusion in each experiment are detailed in each chapter.

DISSERTATION STRUCTURE

This dissertation tests the effects of three of the most important processes in seedling

recruitment. Chapter 2 broadly compares the degree to which dispersal and environmental condi-

tions shape the density, diversity, and species composition of the seedling layer. This comparison

provides a critical test of two main theories of species coexistence: niche-differentiation and

neutrality. Evidence that the relatively deterministic processes that occur during seedling estab-

lishment (i.e., ecological filtration) govern seedling recruitment would suggest that theories of

17

Table 1.1. Characteristics of the 18 species used in this dissertation. Seed mass, in grams, is presented ± 1 SD. Adult density is a single estimate, and is therefore presented without an estimate of variance. “Experiments Used” indicates in which experiments each species was used: “Dispersal” = Chapter 2, “Competition” = Chapter 3, “Predation” = Chapter 4. Details on the estimation of seed mass and adult abundance are provided in Chapter 4. “Germination” indicates the percentage of seeds observed to germinate in a greenhouse experiment. It is presented with standard deviation and sample size. Ten seeds were sown into each pot, and observed to germinate and grow over 18 months. “nd” indicates no data were available.Seed and seedling images are extracted from http://manuplants.org, where they and many others are freely available. Pouteria reticulata, used in the Panamanian component of the competition experiment (Chapter 3), is not pictured. All scale bars are graduated in millimeters.

http://manuplants.org

18

��

Bonpl.

(Bombaceae)

5.29 ± 0.78

Storage, epigeal

7.4

nd

Competition

��

Cambess.

(Clusiaceae)

2.52 ± 0.77

Storage, hypogeal

0.4

56 ± 13 (10)

Predation

��

Ruiz & Pavon

(Arecaceae)

3.86 ± 0.41

Storage, hypogeal

32.2

54 ± 22 (10)

Dispersal, Predation

0.61 ± 0.14

Storage, epigeal

0.2

33 ± 23 (10)


0.41 ±�0.07

Storage, epigeal

3.6

55 ± 24 (4)


0.79 ± 0.06

Storage, epigeal

2.3

45 ± 15 (10)


1.93 ± 0.49

Photosynthetic

0.2

42 ± 13 (10)


3.44 ± 0.56

Storage, hypogeal

0.1

77 ± 16 (10)


��

Hiern.

(Ebenaceae)

��

(A. DC.) J.F. Macbr.

(Ebenaceae)

��

Ducke

(Combretaceae)

��

(Mart.) H. Wendl.

(Arecaceae)

��

Benth.

(Annonaceae)

Scientific Name

��

1.78 ± 0.88

Storage, hypogeal

3.8

86 ± 12 (10)

Predation

0.49 ± 0.15

Photosynthetic

0.4

56 ± 13 (10)

Predation

0.22 ± 0.07

Storage, epigeal

0.3

55 ± 10 (10)


0.55 ± 0.10

Photosynthetic

~0.1

13 ± 15 (10)

Predation

0.61 ± 0.07

Photosynthetic

0.2

70 ± 12 (10)

Predation

1.80 ± 0.16

Storage, epigeal

37.3

20 ± 12 (10)

Predation

1.11 ± 0.75

Storage, hypogeal

1.3

nd

Competition

0.05 ± 0.003

Photosynthetic

0.9

65 ± 19 (10)

Predation

0.86 ± 0.16

Storage, epigeal

3.4

23 ± 12 (6)

Dispersal

��

Bertero ex Spreng.

(Verbenaceae)

��

L.

(Rubiaceae)

��

Engl.

(Olacaceae)

��

(Markgr.) A.H. Gentry

(Myristicaceae)

��

Ruiz & Pavon

(Moraceae)

��

Sw.

(Moraceae)

��

(A. Juss.) C. DC.

(Meliaceae)

��

��

Meisn.

(Hernandiaceae)

�� sp. nov.

(Flacourtiaceae)��

��

��

��

��

��

Table 1.1.

19

��

Bonpl.

(Bombaceae)

5.29 ± 0.78

Storage, epigeal

7.4

nd

Competition

��

Cambess.

(Clusiaceae)

2.52 ± 0.77

Storage, hypogeal

0.4

56 ± 13 (10)

Predation

��

Ruiz & Pavon

(Arecaceae)

3.86 ± 0.41

Storage, hypogeal

32.2

54 ± 22 (10)


0.61 ± 0.14

Storage, epigeal

0.2

33 ± 23 (10)


0.41 ±�0.07

Storage, epigeal

3.6

55 ± 24 (4)


0.79 ± 0.06

Storage, epigeal

2.3

45 ± 15 (10)


1.93 ± 0.49

Photosynthetic

0.2

42 ± 13 (10)


3.44 ± 0.56

Storage, hypogeal

0.1

77 ± 16 (10)


��

Hiern.

(Ebenaceae)

��

(A. DC.) J.F. Macbr.

(Ebenaceae)

��

Ducke

(Combretaceae)

��

(Mart.) H. Wendl.

(Arecaceae)

��

Benth.

(Annonaceae)

Scientific Name

��

1.78 ± 0.88

Storage, hypogeal

3.8

86 ± 12 (10)

Predation

0.49 ± 0.15

Photosynthetic

0.4

56 ± 13 (10)

Predation

0.22 ± 0.07

Storage, epigeal

0.3

55 ± 10 (10)


0.55 ± 0.10

Photosynthetic

~0.1

13 ± 15 (10)

Predation

0.61 ± 0.07

Photosynthetic

0.2

70 ± 12 (10)

Predation

1.80 ± 0.16

Storage, epigeal

37.3

20 ± 12 (10)

Predation

1.11 ± 0.75

Storage, hypogeal

1.3

nd

Competition

0.05 ± 0.003

Photosynthetic

0.9

65 ± 19 (10)

Predation

0.86 ± 0.16

Storage, epigeal

3.4

23 ± 12 (6)

Dispersal

��

Bertero ex Spreng.

(Verbenaceae)

��

L.

(Rubiaceae)

��

Engl.

(Olacaceae)

��

(Markgr.) A.H. Gentry

(Myristicaceae)

��

Ruiz & Pavon

(Moraceae)

��

Sw.

(Moraceae)

��

(A. Juss.) C. DC.

(Meliaceae)

��

��

Meisn.

(Hernandiaceae)

�� sp. nov.

(Flacourtiaceae)��

��

��

��

��

��

Table 1.1, continued

20

niche-differentiation best explain community structure and dynamics, at least at the seedling

stage. On the other hand, evidence that the stochastic process of dispersal limits seedling recruit-

ment would indicate that theories of neutrality may provide a better explanation.

The focus of Chapter 2 is (intentionally) too broad to dissect the relative importance of

the various processes that occur during seedling establishment. The experiments of the two sub-

sequent chapters are designed to provide additional detail about the sources of mortality faced

by establishing seedlings. Chapters 3 tests the effects of competition among seedlings. While the

asymmetrical competitive effects of adult trees on seedling recruitment are powerful and well-

documented, the potential for competition among seedlings has not been explored before now. If

intense, inter-seedling competition would impose an additional filter that could limit seedling

recruitment.

Chapter 4 tests the effects of mammalian predation on seedling recruitment. Seed preda-

tion and seedling herbivory have long been known to reduce the density of recruited seedlings.

I compare the effects of small, medium and large mammals as seed predators and seedling

herbivores. I also assess the effects of each mammal size-class on seedling diversity. Chapter 4

has been accepted for publication in Ecology, and is henceforth referred to as Paine and Beck (in

press).

In Chapter 5, I revisit the stages of seedling recruitment reviewed here to integrate the

results from my experiments with previously published studies. Combining my results with those

of other investigators, I attempt to identify those factors that are most important in generating

variation in seedling community structure.

21

CHAPTER 2 — WHAT SHAPES TROPICAL SEEDLING COMMUNITY STRUCTURE? SEED DISPERSAL VERSUS

ENVIRONMENTAL CONDITIONS

INTRODUCTION

Two bodies of theory, niche-differentiation and neutrality, attempt to explain the coexis-

tence of species in diverse habitats. Niche-differentiation theories suggest that diversity results

from relatively deterministic tradeoffs and interactions among species that play out across spa-

tially heterogeneous landscapes (Chase and Leibold 2003). Niche-differentiation theories make

three predictions that have found strong empirical support. The species composition of commu-

nities and the spatiotemporal distribution of individuals should track environmental conditions

(Potts et al. 2002, Clark and McLachlan 2003, John et al. 2007). Community composition should

also be structured by competitive and predatory interactions among individuals (Fargione et al.

2003, Harpole and Tilman 2006, Paine and Beck in press). Differences in traits among species

should translate into tradeoffs in demographic performance (Kitajima 1994, Bloor and Grubb

2003). However, neutral theories predict macroecological patterns just as accurately as niche-

based theories, and much more parsimoniously (McGill et al. 2006).

Neutral theories take the view that all individuals are ecologically equivalent, experi-

encing precisely equal likelihoods of growth and mortality (Bell 2001, Hubbell 2001). Neutral

theories suggest that diversity arises through a dynamic interplay of each species’ stochastic (but

inexorable) walk to extinction, and regional immigration from the metacommunity (McGill et

al. 2006). While challenged by abundant evidence of non-neutral patterns, neutral theorists draw

attention to the ability of their theories to predict accurately and parsimoniously macroscopic

22

aspects of community structure, such as relative species abundance (Hubbell 2001, Volkov et al.

2003, Volkov et al. 2005).

The two bodies of theory disagree as to how the stages of recruitment affect the density,

diversity and species composition of plant communities. The recruitment of seedlings from seeds

comprises two stages, dispersal and establishment (Nathan and Muller-Landau 2000, Muller-Lan-

dau et al. 2002, Chapter 1). The dispersal of seeds is a stochastic, distance-limited process that

may limit the recruitment of individual species (reviewed by Turnbull et al. 2000). Neutral theo-

ries take the view that stochastic processes, especially seed dispersal, structure plant communities,

and that inter-specific trait variation has no bearing on demography. Thus, in a fully neutral forest,

the spatiotemporal distribution of individuals generated by dispersal would persist throughout

establishment, because the mortality of individuals would occur randomly with respect to species.

Nevertheless, a vast body of work demonstrates that environmental conditions, in con-

junction with inter-species trait variation, affect seedling growth and survival. In tropical forests,

these conditions include: soil moisture (Engelbrecht and Kursar 2003), light availability (Nicotra

et al. 1999, Montgomery and Chazdon 2002), competition (Lewis and Tanner 2000, Chapter 3)

and seed predation (Paine and Beck in press), among others. I refer to ‘environmental conditions’

in the broad sense, including all factors, biotic and abiotic, that potentially affect seedling estab-

lishment. Niche-differentiation theories emphasize the degree to which these factors shape plant

community structure. To the extent that empirically measured variables represent variation rel-

evant to seedlings, niche-differentiation suggests that variation in environmental conditions should

predict community structure. Accordingly, the observation that dispersal-generated distribution

of seedlings is subsequently altered by variance in environmental conditions would be consistent

with niche-differentiation theories.

23

I tested the effects of seed dispersal and environmental conditions on the stem den-

sity, species diversity, and species composition of tree seedlings with a series of seed-addition

experiments in a highly diverse Amazonian rainforest. The two bodies of theory make clear and

contrasting predictions. In short, neutrality predicts that seed dispersal structures plant communi-

ties, whereas niche partitioning predicts that environmental conditions do so. It is increasingly

evident that both theories apply in all habitats, though in differing proportions (Chase 2003). I

therefore adopted a variance-partitioning approach to weigh the support for these predictions

from observations of the structure of the seedling layer. This statistical approach, coupled with

manipulations of seed dispersal, allowed me to conduct an unusually strong test of the theories of

neutrality and niche-differentiation (level D3 sensu McGill et al. 2006). Furthermore, for a cohort

of individuals, the relationship between community structure and environmental conditions may

be expected to intensify with time since germination. I therefore examined how the importance of

each body of theory changed during the two years after seeds were sown.

METHODS

Study Site and Species

This study was conducted at Estación Biológica Cocha Cashu (EBCC) in the lowland

floodplain rainforest of Peru’s Manu National Park; see detailed site description in Chapter 1.

Species were included in the seed-addition experiments of this study based upon three

criteria. First, their seeds had to be sufficiently large to be easily cleaned and sown. Second, fruit

needed to be available between April – May 2004, the period of seed additions. And third, fruit-

ing adults had to be sufficiently common and fecund to provide 1600 – 3100 seeds per species for

sowing. These criteria yielded 8 species, representing 6 families, all of which are primarily mam-

mal-dispersed (Foster and Janson 1985). The focal species include six trees and two palms, all of

24

which reach the canopy as adults. Adult abundance and seed masses were determined as in Paine

and Beck (in press). Adult abundance of the 8 species ranged from 0.08 – 32 trees/ha (median =

1.3 trees/ha), and their seed mass ranged from 0.4 – 3.9 g (median = 0.8 g). Species names and

distinguishing characteristics are presented in Tables 1.1 and 2.1.

Experimental Design

We used seed-addition experiments to test the roles of dispersal and environmental varia-

tion in shaping seedling community structure. In July 2003, 230 plots were laid out in 23 blocks

of 10 plots each. The following April, each block was expanded to 24 plots, and a 24th block was

added for a total of 576 study plots. Blocks were randomly located throughout a 4-km2 area of

mature floodplain forest at EBCC (See Figure 1.2). Blocks were located to avoid recent tree-fall

gaps or frequently inundated swales. Within a block, plots were arranged in two (occasionally

three) parallel rows, separated by 10 m. Spacing among plots within rows varied between 5 and

10 m to minimize autocorrelation with environmental variables. The geographic distance among

plots ranged from 5 m to 2 km. Plots were circular, with an area of 1 m2, and each was perma-

nently marked with a central iron rebar. Over the three-year study, four of the 576 plots were

destroyed by fallen trees, and were excluded from all analyses.

Density, diversity, and control treatments were randomly interspersed to plots within

blocks. Each block contained 16 density plots, 4 diversity plots, and 4 control plots. The 16 den-

sity plots in each block were, in turn, divided evenly among the eight focal species. In density

treatment plots, seeds of a single species were sown at doubling classes of density up to a species-

specific maximum. During the fruiting season, I exhaustively searched 10 randomly located 1-m2

quadrats beneath the crowns of at least five fruiting adult trees to determine the maximal density

of the natural seed rain. The maximum observed density ranged among species from 48 seeds/m2

25

for Duguetia quitarensis and Diospyros subrotata to 120 seeds/m2 for Iriartea deltoidea, and

was used as the maximal sowing density (Table 2.1). Diversity treatment plots received exactly 24

seeds, representing 2, 4, 6, or 8 species, with species drawn randomly and evenly from the pool

of eight focal species. In this way, evenness was maximized, and the species richness of the sown

seeds was confounded neither with sown density nor species composition. Annually, 1-m2 quad-

rats of forest floor receive seeds of a median of 6 vertebrate-dispersed species (quartiles: 4 – 8),

according to Dr. John Terborgh’s three-year seed-trapping project at EBCC (J. Terborgh, personal

communication). My diversity seed-additions thus approximated the range of species richness

Species (Family)Annual Fruiting Peak

Seed rain (seedsm-2 y-1)

Maximum sown density (seeds/m2)

Adult density (ha-1)

Seed mass (grams)

Duguetia quitarensis Benth. (Annonaceae) April 0.36 48 3.60 0.41

Iriartea deltoidea Ruiz & Pav. (Arecaceae) January 0.24 120 32.30 3.86

Socratea exorrhiza (Mart.) H. Wendl. (Arecaceae) February < 0.01 56 0.08 3.44

Buchenavia grandis Ducke (Combretaceae) March < 0.01 96 0.17 1.93

Diospyros pavonii (A. DC.) J.F. Macbr.(Ebenaceae) August 0.02 60 0.17 0.61

Diospyros subrotata Hiern. (Ebenaceae) June < 0.01 48 2.33 0.79

Trichilia pleeana (A. Juss.) C. DC. (Meliaceae) February 0.13 80 3.35 0.86

Heisteria nitida Engl. (Olacaceae) April 0.06 64 0.25 0.22

Table 2.1. Species sown in the seed-addition project, sorted by family. All species are canopy trees, except Iriartea deltoidea and Socratea exorrhiza, which are canopy palms. All species have moderately reliable annual fruiting phenology. Annual seed rain was estimated from the seed rain into the 289 1-m2 seed traps of Dr. John Terborgh between January 2003 and January 2006. Seeds were sown in four density treatments, from one-eighth of the maximum and increasing by doubling classes to the maximum (shown here). Seed mass is the air-dry weight of fresh seeds with all dispersal structures removed.

26

naturally occurring in the seed rain. Unsown plots were used as controls for both experiments.

Seedlings of focal species germinating in the control plots estimated the natural recruitment rate.

Seeds were manually sprinkled over plots without disturbing the soil in any manner (as

in Tilman 1997). The vegetation in each plot was lightly shaken to ensure that seeds settled to

the soil surface. The density plots for Diospyros pavonii were sown in July 2003 as a pilot study.

Density plots for the other seven species, and all diversity plots, were sown in May 2004.

After sowing, plots were observed every six months between July 2003 and July 2006,

at the peaks of the dry season (July) and wet season (January). At every census, I tagged all

seedlings of the eight focal species as they germinated. In addition to observing the eight focal

species, I monitored three aspects of the environment of each plot: the biotic neighborhood, soil

moisture and canopy openness. I evaluated two components of each plot’s biotic neighborhood. At

every July census, I tagged and measured seedlings and small saplings of free-standing non-focal

woody species between 10 and 50 cm tall, excluding any that bore only cotyledons. I define a

seedling as having recruited if it met these criteria. At all censuses except in the wet seasons of

2004 and 2006, I estimated the percent cover of understory palms, ferns, herbaceous vegetation

(mostly Marantaceae, Heliconiaceae, and Araceae), and coarse woody debris (limbs, trunks and

roots > 5 cm in diameter). I used the four estimates of cover, along with the stem density and spe-

cies richness of non-focal seedlings as descriptors of the biotic neighborhood. Soil moisture was

evaluated gravimetrically to a depth of 5 cm in July 2004, and with a time-domain reflectometer

in July 2005, January 2005 and January 2006 (depth: 15 cm; Field Scout TDR 200, Spectrum

Technologies). Light availability was quantified as Global Light Index in Gap Light Analyzer 2.0

(Frazer et al. 2000) from canopy photos taken 67 cm above each plot in July 2004.

27

Identifying small seedlings in such a diverse forest is challenging. To facilitate identifica-

tion, I developed living and digital reference collections of seedlings. Seeds of all identifiable

fruits were sown into a greenhouse at EBCC between 2003 and 2005. Greenhouse-grown

seedlings were studied and photographed repeatedly throughout their germination and establish-

ment (all images and information publicly available at http://manuplants.org). All seedlings were

identified in the field, based on field characteristics and comparisons with greenhouse-grown

individuals. Voucher photographs of unidentified seedlings were taken in the field. Post-hoc com-

parisons of voucher photographs with the reference collection allowed further identifications to

be made. Nevertheless, not all seedlings could be confidently identified to species. Recognizable

seedlings that could not be assigned to a species were assigned to higher taxa, and/or assigned to

morphotypes. I use the term “species” synonymously with “morphotype”. The identification of

each seedling was re-evaluated and refined (if necessary) at each census. By the final census, in

the dry season of 2006, 3% of seedlings could not be confidently assigned to family; 75% and

93% were identified to species and genus, respectively. To increase taxonomic consistency and

minimize any potential bias stemming from misidentifications, analyses of diversity and species

composition were conducted at the genus level.

This study broadly compared the effects of dispersal and environmental conditions. I

experimentally manipulated the former, and not the latter. The vast majority of seeds dispersed

in tropical rainforests land in shaded understory conditions, as recent tree-fall gaps and other

potentially high-resource sites are relatively scarce (1-2% by area, Lieberman et al. 1985, H. Beck,

unpublished data). Thus, this study examined how variation in seed rain density and species rich-

ness affect seedling community structure in the context of the natural variation in environmental

conditions encountered in the shaded understory of tropical rainforests. Observed variation in

http://manuplants.org

28

environmental conditions nevertheless spanned the range expected in the understory of a neotropi-

cal rainforest (Denslow et al. 1991, Nicotra et al. 1999, Montgomery and Chazdon 2002, Harms et

al. 2004, Chapter 3). It is reasonable to compare manipulated seed dispersal and unmanipulated

environmental conditions, because even relatively minor changes in environmental conditions

can significantly affect seedling growth and survival (e.g., Montgomery and Chazdon 2002).

Moreover, manipulation of dispersal treatments and multiple environmental conditions in a facto-

rial manner would have resulted in low replication of sample units.

Data Analyses

I assessed the variance explained by dispersal treatments, environmental variables, and

their one-way interactions in terms of their effects on four aspects of community structure: stem

density, species richness, Shannon-Weiner diversity, and species composition. I assessed their

effects on three nested sets of seedlings: seedlings of focal species, all seedlings (both focal and

non-focal) that recruited exclusively between July 2003 and July 2004 (henceforth ‘cohort of

2004’), and the entire seedling layer together (regardless of their age).

Correlations between establishment conditions or dispersal treatments and geographic

distance could confound their relationships with seedling community structure. Accordingly, I

assessed the correlation between environmental conditions and geographic distance with Mantel

tests (Mantel 1967, Gilbert and Lechowicz 2004). The correlation between two distance matrices

cannot be evaluated with standard methods because of the lack of independence of values within

each matrix (McCune and Grace 2002). Mantel tests are nonparametric, using randomization of

one distance matrix to assess the significance of the observed correlation. First I calculated the

Euclidian distances among plots, and the scaled Euclidian dissimilarity of environmental condi-

tions of each plot. From these distance matrices, I calculated the Pearson correlation coefficient.

29

Then I permuted the elements of one matrix, recalculated the correlation, and determined how

often a stronger correlation would be obtained by chance. Significance of observed correlations

were assessed with 1000 permutations. Correlations between environmental variables and geo-

graphic distances were very weak at all census periods (Mantel’s r ≤ 0.087), but were significant

in July 2004, January 2005 and July 2005 (P ≤ 0.04, Figure 2.1). Dispersal treatments were ran-

domly assigned to plots, and should, therefore, have been uncorrelated with geographic distance.

I tested this assumption with another set of Mantel tests, and found it to be valid (Mantel’s r ≤

0.020, P ≥ 0.48, data not shown).

To assess the relative importance of dispersal treatments and environmental variables

for log-transformed seedling density and species richness, I used stepwise model selection based

on Akaike’s Information Criterion (AIC, Burnham and Anderson 2002). I began with a multiple

regression that included log-transformed sown density (or log-transformed sown species rich-

ness) and all measured environmental variables as predictor variables. Stepwise selection was

used to select the best combination of variables (including one-way interactions) predicting each

aspect of community structure. Models were compared on the basis of AIC, and both forward

and backward steps were possible (Venables and Ripley 1999). Separate models were created for

each census period. Repeated measures analyses were not conducted, as not all environmental

conditions were measured at all census periods. I extracted the percent variance explained by

dispersal treatment, and the sum of the variance explained by establishment variables, and by dis-

persal-establishment interactions, if any were chosen in the stepwise selection, using Type I sums

of squares. Because dispersal is the first stage of recruitment, prior to establishment, dispersal

treatment always entered the model first. Thus, the percent variance explained by environmental

conditions represented that which remained after accounting for dispersal.

30

Figure 2.1. Environ-mental similarity was weakly and inversely, but significantly, correlated with geographic distance. Points indicate the envi-ronmental distance among all pairs of the 572 plots. Line indicates the lowess-smoothed local regression relationship, and sug-gests that the relationship between environmental similarity and geographic distance among plots was very weak.

31

Finally, I used another set of Mantel tests to test the (partial) correlations between sown

species composition, environmental conditions, and seedling species composition (McCune and

Grace 2002). The correlation between sown and seedling species composition was calculated as

described above. The relationship between environmental conditions and seedling species compo-

sition was calculated as a partial correlation coefficient. As in the seedling density and diversity

analyses, this approach allowed an estimation of the partial correlation of environmental condi-

tions and species composition, after accounting for the correlation of sown and seedling species

composition (Smouse et al. 1986, Castellano and Balletto 2002). Distance matrices of species

composition were calculated with Sorenson (Bray-Curtis) dissimilarity, which incorporates both

presence-absence and density data. I tested the significance of the (partial) correlations by ran-

domly permuting the seedling species composition matrix 1000 times (holding constant the sown

species composition and environmental similarity matrices), and recalculating the correlation

coefficient. Mantel tests were repeated at each census period to assess how the relative importance

of dispersal and establishment changed over time.

All analyses were performed in the R statistical package (R Development Core Team 2006

version 2.4.1). The MASS and vegan packages were used to implement the stepwise selection and

Mantel tests, respectively (Venables and Ripley 1999, Oksanen et al. 2005).

RESULTS

Over three years, 1360 seedlings of focal species germinated from the 15,132 seeds that

I had sown; constituting a mean germination rate of 9.0%. Germination varied among species,

from 2.1% in Duguetia quitarensis to 21.3% in Diospyros pavonii. 106 seedlings of focal spe-

cies recruited in control plots to which no seeds were experimentally added. This translates to

a natural recruitment rate of 0.06 seedlings m-2 year-1 of focal species. Three relatively common

32

Stem

Density

A

SpeciesRichness

BFocal Non-focal Cohort of 2004

0

3

6

9

0

2

4

6

8

July 2003 July 2004 July 2005 July 2006

Figure 2.2. (a) Stem density and (b) species richness of focal and non-focal seedlings over three years. Focal seedlings made up a small portion of the overall seedling layer. Those individuals and species that comprise the cohort of 2004 are outlined in black. Note that the cohort of 2004 consisted of both focal and non-focal species.

33

and fecund species recruited more than 10 seedlings into control plots, whereas no seedlings of

Socratea exorrhiza, which is rare in the local area, recruited naturally.

I tagged 4792 non-focal seedlings over the three-year study, representing 320 species, 185

genera and 66 families (see Appendix for details). Each 1-m2 study plot contained, on average, 6.4

± 4.8 individual stems, and 5.0 ± 3.0 species (mean ± SD; Figure 2.2). Non-focal seedling density

peaked in July 2004, and declined thereafter. I did not monitor natural seed rain in the context of

this study, but this degree of temporal variation in seedling stem density has been associated with

fluctuations in seed rain density elsewhere in Amazonia (Norden et al. 2007).

Dispersal was a better predictor of seedling density than were environmental conditions

(Figures 2.3 & 2.4). For seven of eight species, sown density explained more of the variance in

seedling density than did all environmental variables combined, indicating that dispersal strongly

limits seedling density. For the eighth species, Duguetia quitarensis, seedling stem density was

low and unrelated to sown density, suggesting that establishment, more than dispersal, limits

seedling stem density. Model fits were strong, for ecological data; models of seedling stem density

explained an average of 39% of total variance. Not surprisingly, the percent variance explained

by sown density was initially substantial (mean over 8 focal species, January 2005: 27%), and

declined by the end of the study to 16%. Nevertheless, at the final census, dispersal still signifi-

cantly limited stem density, explaining 6 – 41% of the variance. At the final census, establishment

explained more variance in stem density for three species than did sown density. Statistical

interactions among dispersal treatments and environmental variables were frequently significant,

but rarely explained more than 10% of the observed variance. Taking all eight species together,

experimental manipulations of sown density explained only a small fraction of the variance in the

cohort of 2004 stem density (which included both focal and non-focal species, Figure 2.4). Instead,

34

0

2

4

6

0

2

4

6

0

4

8

0.0

0.2

0.4

0.6

0.00.51.01.5

0

1

2

0.0

0.4

0.8

0

4

8

Seedlingstem

density

Observation Period

Sown Density4: Maximum321: Minimum0: Control

Buchenavia grandis Diospyros pavonii

Diospyros subrotata Duguetia quitarensis

Iriartea deltoidea

Socratea exorrhiza Trichilia pleeana

Heisteria nitida

Figure 3

Jul2003

Jan2004Jul Jan

2005Jul Jan

2006Jul Jul

2003Jan2004Jul Jan

2005Jul Jan

2006Jul

Figure 2.3. Density of recruited seedlings of eight species over time, by sown density. Adding a dense seed rain increased seedling density for seven of eight species. Maximum sown density varied among species - see Table 2.1. Arrows indicate the timing of seed-additions. Note that the scales of the y-axes differ.

35

Figure 2.4. The variance in focal species seedling stem density was better explained by seed dispersal than by environmental conditions. Each bar represents the percent variance in seedling density explained by dispersal treatment, environmental conditions, their interactions, or unexplained. Bars are grouped by species, and sorted by census period within species. The lowermost 2 sets of bars represent the effects of sown density on the density of the cohort of 2004 and overall density, respectively. The percent variance explained by environmental conditions represents the sum of the variance explained by all establishment variables summed together. Only species that germinated rapidly upon sowing were analyzed in July 2004; “nd” indicates that no data were available for analysis. “***” indicates P < 0.0001, “**” indicates P < 0.001, and “*” indicates P < 0.01.

36

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

PercentVariance Explained

Jul 04Jan 05Jul 05Jan 06Jul 06








Jul 04Jul 05Jul 06

Jul 04Jul 05Jul 06

Buchenavia grandis

Diospyros pavonii

Diospyros subrotata

Duguetia quitarensis

Heisteria nitida

Iriartea deltoidea

Socratea exorrhiza

Trichilia pleeana

Cohort of 2004stem density

Overallstem density

*** ** ****** ***** ******** ***

*** *** ***** **** *** ******** *** ***

nd*** * **** * ****** * ****** * ***

nd*** *** ***

nd*** ***** ******* * **

nd*** ** ***** ** ******** **

nd*** **** * *******

* ***** ***** ***** ******** *** *

*** ****** *** **** ***

*****

Dispersal

Establishment

Interaction

Residual

PercentVarianceExplained by:

Figure 4

Figure 2.4

37

0

1

2

Sown Richness8 species6420 species

Focal Species

0

1

2

Seedlingspeciesrichness

Cohort of 2004

2

4

6

8

Observation period

Overall

Figure 5

Jul2003

Jan2004

Jul Jan2005

Jul Jan2006

Jul

Figure 2.5. Species richness of focal species, the cohort of 2004, and all seedlings together. Each line represents the species richness in plots where 0, 2, 4, 6, or 8 species were added as seeds in July 2004. Adding a diverse seed rain significantly increased focal seedling species richness, but had limited effects on cohort richness and overall richness, since their inter-plot variance was great. Arrows indicate the timing of seed additions. Note that the scales of the y-axes differ.

38

the stem density of the cohort of 2004 was best explained by environmental variables (R2 > 0.66).

Overall seedling stem density was well explained neither by sown density nor environmental con-

ditions (R2 < 0.09).

The species richness of focal seedlings was strongly dispersal-limited, but was also influ-

enced by environmental variation (Figures 2.5 & 2.6). In January 2005, sown species richness

explained 27% of the variance in seedling species richness; environmental variables explained 2%.

Eighteen months later, at the final census, the variance explained by dispersal and environmental

conditions had roughly equalized, at 8% and 11%, respectively. The results for analyses of Shan-

non-Weiner diversity closely paralleled those of species richness, and are not shown. The species

richness of the cohort of 2004 was well explained by environmental variables (R2 > 0.42), and not

by sown species richness (R2 < 0.03). Overall seedling species richness was well explained neither

by dispersal treatments, nor environmental variables (R2 < 0.06).

Seedling species composition was more tightly correlated with sown species composition

than with environmental conditions (Table 2.2). The correlation between sown species composi-

0% 20% 40% 60% 80% 100%

Percent Variance Explained


Jul 04Jul 05Jul 06

Jul 04Jul 05Jul 06

Focalspecies richness

Cohort of 2004species richness

Overallspecies richness

nd*** ********** *

*********

* **

Figure 6

Figure 2.6. Variance in species richness of focal seedlings, the cohort of 2004, and the overall seedling layer was explained by seed dispersal, environmental conditions, and their interactions. Only three focal species had germinated by July 2004, so the species richness of focal seedlings was analyzed only after January 2005. Abbreviations are as in Figure 2.4.

39

Dispersal alone Establishment after dispersal r P r P

Focal SpeciesJuly 2004 nd nd nd ndJanuary 2005 0.33 < 0.001 0.00 0.585July 2005 0.25 < 0.001 0.02 0.134January 2006 0.25 < 0.001 0.03 0.019July 2006 0.22 < 0.001 -0.01 0.603

Cohort of 2004July 2004 0.40 < 0.001 0.01 0.319July 2005 0.36 < 0.001 0.06 0.006July 2006 0.33 < 0.001 0.02 0.177

Overall communityJuly 2004 0.11 < 0.001 0.08 0.019July 2005 0.08 < 0.001 0.10 < 0.001July 2006 0.17 < 0.001 0.04 0.038

Table 2.2. Dispersal, not establishment, predicts seedling species composition, even 2+ years after a one-time addition of seeds. The effect of seed dispersal on seedling species composition was assessed with a Mantel test of sown species composition versus seedling species composition. The effect of post-dispersal establishment variables was assessed with a partial Mantel test of establishment variables versus seedling species composition, while controlling for sown species composition. “nd” = no data.

tion and seedling species composition was strongest in January 2005 (Mantel’s r = 0.33), and

weakened over the following two years to 0.22. Partial correlations between environmental con-

ditions and focal species composition, after accounting for dispersal, were very weak (r ≤ 0.03),

and were only significant in January 2006 (P = 0.019). The species composition of the cohort of

2004 was more strongly and significantly correlated with sown species composition than with

environmental conditions. Only in July 2005 was cohort species composition correlated with

environmental conditions (P = 0.006). Overall seedling species composition was more tightly

correlated with sown species composition than with environmental variables (Mantel’s rdispersal:

0.08–0.17, Mantel’s renvironment: 0.04–0.10).

40

DISCUSSION

The importance of niche and neutral theories in any given community may be stringently

assessed through experiments in which each body of theory generates clear, contrasting predic-

tions, coupled with a statistical framework that permits an estimation of the degree of support for

each (McGill et al. 2006). This study provides such an approach. Neutrality predicts that seed

dispersal will structure the seedling layer, whereas niche-differentiation predicts that variation in

environmental conditions will do so. Though stochasticity in any demographic process may affect

community structure, seed dispersal is considered a major source of the stochasticity required for

neutral dynamics (Hubbell 2001). This variance-partitioning approach allowed me to assess the

degree to which each prediction explains my observations of the seedling layer.

Dispersal Limits Density and Diversity, and Shapes Species Composition

My results show that dispersal limits the density of tree seedlings in a highly diverse

Neotropical rainforest. This outcome is consistent with the results of two recent seed-addition

studies conducted in Kenyan and Panamanian rainforests (Makana and Thomas 2004, Svenning

and Wright 2005). We further found that the species richness of the seed rain also limits seedling

species richness, a novel result in tropical rainforest. This finding corroborates results from (often

highly manipulated) temperate grasslands (Tilman 1997, Foster and Tilman 2003, Foster et al.

2004). It is also consistent with tropical rainforest studies that have inferred dispersal limitation

to be strong, either through seed trapping (Clark et al. 1998, Clark et al. 1999, J. Terborgh unpub-

lished data), or through observations of adult distributions at local (Seidler and Plotkin 2006) and

regional scales (Condit et al. 2002, Potts et al. 2002, Tuomisto et al. 2003).

My results further demonstrate that a single seed-dispersal event can shift the species

composition of the seedling layer. This finding is robust, in that the species composition of

41

experimentally dispersed seeds was positively and significantly correlated with the composition of

focal species seedlings, the cohort of 2004 seedlings, and the entire seedling layer (Table 2.2). In

this diverse forest, the arrival of a seed of a rare species could substantially shift a plot’s species

composition. My 572 plots contained a total of 320 species, yet the average species richness per

1-m2 plot was 5.0 species. While my seed additions included species that were common, they also

included some species whose seedlings very rarely recruited naturally into control plots, because

of the rarity of adult trees, limited seed dispersal, or both (Table 2.1). Thus my seed additions

were sufficient to shift the species composition of the seedling layer. In less diverse sites, dispersal

may have less of an impact on seedling species composition than in highly diverse sites. Diverse

sites contain a large proportion of rare taxa. In less diverse habitats, a plot’s species composi-

tion would be less affected by the arrival of any given seed, because on average, each plot would

already contain a greater fraction of the species pool than would plots in highly diverse habitats.

Relative Importance of Neutrality and Niche-Differentiation

A persistent question in the niche-neutrality debate is ‘…“under what conditions is one

model better than another in explaining the data?”’ (McGill et al. 2006). It is increasingly recog-

nized that the relative importance of neutrality and niche-differentiation varies continuously, both

within and among sites. This recognition may help to resolve the debate between partisans of each

theory (Leibold and McPeek 2006).

Four factors are likely to be important in determining the relative importance of the two

bodies of theory: the intensity of ecological filtration, the scale of analysis, the aspect of commu-

nity structure analyzed, and the time since dispersal occurred. The degree to which environmental

conditions (biotic and abiotic) filter individuals from the community may correlate with the

degree to which niche-differentiation structures the community. Sites with harsh environments,

42

intense predation or intense disturbance regimes – i.e., sites with intense ecological filtration – are

particularly resistant to stochasticity in community structure (Chase 2003, Fukami 2004a, b).

These sites are likely to have relatively low diversity, relatively predictable species compositions,

and may be better-described by niche-partitioning theories of species coexistence (McGill 2003,

McGill et al. 2006). Ecological filtration may also limit the distributions of co-occurring species

with a community to differing degrees (Moore and Elmendorf 2006).

On the other hand, where the processes of ecological filtration are relatively weak, neutral

dynamics may become apparent. Just as neutral drift is expected for a gene not under selection

pressures (Kimura 1983), neutrality may be the default condition for a community in the absence

of ecological filtration. As selection non-randomly alters the relative abundance of alleles, ecologi-

cal filtration may eliminate individuals from a community non-randomly with respect to species.

Ecological filtration may thus drive a community away from neutrality. Sites with relatively

benign environments and moderate rates of predation and disturbance are therefore more likely

to be sensitive to stochasticity in dispersal or arrival order (Chase 2003, Fukami 2004a, b). The

communities that occupy these sites may be well described by neutral theories of coexistence.

Interestingly, within-site diversity per se may not be directly related to the applicability

of neutral or niche-differentiation theories. Instead, the applicability of each may be related to

the intensity of ecological filtration. Intense ecological filtration often causes sites to exhibit low

diversity. Because ecological filtration typically favors specific clades, or taxa that display specific

traits (Harpole and Tilman 2006, McGill et al. 2006), the species that coexist in intensely filtered

sites tend to be more similar (either phylogenetically, or in terms of traits, or both) than would be

expected by chance. Species-poor sites are often those in which ecological filtering is intense, and

the communities that occupy such sites are often well described by theories of niche-partitioning.

43

The relative importance of niche partitioning and neutrality is also likely to vary with the

scale of investigation. Even if the diversity of the species pool is held constant, larger plots are

more resistant to stochastic perturbations resulting from seed dispersal because they contain more

individuals and more species (Rosenzweig 1995). Larger plots have greater compositional inertia.

Furthermore, differences in some ecological conditions may become apparent only at relatively

large spatial scales. For example, the diversity of tropical forests is related to annual precipitation,

which varies at regional to continental scales (Gentry 1988). My finding that dispersal affects

structure of the seedling layer at a small (1-m2) scale may fade at larger scales.

The relative importance of neutrality and niche-differentiation may depend on the aspect

of the plant community that is assessed. Most recent studies have examined only species composi-

tion (e.g., Gilbert and Lechowicz 2004, Karst et al. 2005). I examined three aspects of community

structure in this study: stem density, species richness, and species composition. Stem density is

the most labile measure, because dispersal and environmental conditions respectively add and

remove individual stems. Seedlings are constantly recruiting into and (through mortality) leaving

the seedling layer. The pattern of stem density generated by seed dispersal is rapidly overprinted

by patterns induced by environmental conditions. Thus, the temporal duration of significant

effects of dispersal on stem density are limited. Species richness is more resistant to stochastic-

ity in seed dispersal. The effects of dispersal on seedling species richness should persist beyond

those on stem density, because for environmental conditions to affect species richness, all indi-

vidual stems of a given species must be eliminated. Finally, the signature of dispersal may remain

evident longest in terms of species composition and spatial distribution. Seed dispersal may affect

seedling species composition through the arrival of a single seed of a rare species. In a diverse

forest, the arrival of a single rare species to a plot will make it highly dissimilar from other plots,

44

moving it to a remote part of multi-dimensional “species-composition space”. Because dispersal

sets the initial distribution of individuals, the spatial patterns it forms may be the longest lasting

effects. I did not assess spatial patterns of seedlings in this study, but the spatial signature of dis-

persal can persist and be detected even in the small-scale distributions of adult trees (Seidler and

Plotkin 2006).

Most studies of dispersal limitation have focused on short-lived herbaceous species, which

may attain reproductive maturity during the study’s duration. For long-lived species, seed-addi-

tion experiments usually only study a tiny portion of the life cycle, by tracking a single cohort

through time (e.g., Makana and Thomas 2004, Svenning and Wright 2005). In such studies, the

relative importance of seed dispersal and environmental conditions will depend on the time since

dispersal. For a cohort, dispersal is a singular event that establishes the initial distribution of indi-

viduals and species. Thereafter, the strength of ecological filtering will increase, though perhaps

not linearly, nor even continuously. Strong bouts of ecological filtration, such as those imposed by

fires or cold snaps, may occur at irregular intervals (e.g., Olmsted et al. 1993). In this study, the

variance in seedling stem density and species richness explained by dispersal treatment declined

through time, but the variance explained by environmental conditions did not concomitantly

increase (Figures 2.4 & 2.6). I interpret this to mean that there was substantial variation in envi-

ronmental conditions that were not captured by my data. In contrast, the correlation of sown and

seedling species composition remained strong throughout the study.

These considerations set up interesting comparisons between my study and others that

conducted similarly strong tests of niche-differentiation and neutral theories. Several studies have

aimed to dissect neutral and niche effects on community composition by assessing the correla-

tions of compositional similarity with environmental conditions and with geographic distance

45

(Gilbert and Lechowicz 2004, Karst et al. 2005). This approach is somewhat problematic, because

it depends upon the relationship between species distributions and dispersal limitation. Neutral

theories postulate that dispersal limitation would cause the correlation with geographic distance

to be strong (Condit et al. 2000). Nevertheless, environmental conditions are often spatially

auto-correlated, and could generate similarly strong correlations. Subtle sampling designs and sta-

tistical analyses are required to partition the variance due to neutrality and niche-differentiation in

observational data (Gilbert and Lechowicz 2004). I circumvented this issue by directly manipulat-

ing seed dispersal. I was thus able to test explicitly the core hypothesis of neutral theories, that

stochasticity in seed dispersal generates significant community structure (also see Moore and

Elmendorf 2006). Moreover, this approach allowed me to test not only community composition,

but also stem density and species richness as response variables. Notably, my study assigned a

stronger role to neutral processes than any of these temperate-zone studies (Gilbert and Lechow-

icz 2004, Karst et al. 2005, Moore and Elmendorf 2006). Whether this is due to differences in the

intensity of ecological filtration among sites, or differences in the life-history stages investigated

is not yet clear.

Seed-addition experiments may underestimate the long-term effects of seed dispersal,

regardless of the response variable examined. Most seed-addition experiments pit a one-time seed

addition against ongoing continuously varying environmental conditions (Turnbull et al. 2000).

As in this study, a one-time addition of seeds may not affect the structure of the seedling layer as

a whole. My seed-addition treatments did, however, boost stem density and species richness in the

cohort of 2004. If dispersal shapes most cohorts, as I expect, its effects will accumulate, with sub-

stantial effects of dispersal accruing in terms of community-wide stem density, species richness,

and species composition. In this context, it is important that my one-time seed additions increased

46

the density and diversity of the cohort of 2004, and suggests that seed dispersal may have cumula-

tive effects on later ontogenetic stages. Furthermore, my diversity seed-addition treatments (the

addition of 24 seeds/m2 at varied richness levels) mimicked natural dispersal by frugivorous

monkeys (Russo and Augspurger 2004). This suggests that seed dispersal, particularly by large

vertebrates, may significantly affect the structure and dynamics of the seedling layer. If the effects

of dispersal persist through later ontogenetic stages, neutral theories may be sufficient to explain

the structure and dynamics of diverse tropical forests.

47

CHAPTER 3 — WEAK COMPETITION AMONG TROPICAL TREE SEEDLINGS: IMPLICATIONS FOR

SPECIES COEXISTENCE

INTRODUCTION

Competition can reduce the growth and survival of individuals, and the sizes of popula-

tions (e.g., Tilman 1994, Pacala et al. 1996). For tropical tree seedlings, whether competition

occurs only between seedlings and adults or additionally among seedlings of similar size has

important ramifications for forest regeneration and the maintenance of species diversity.

If competition is predominantly size-asymmetric, between seedlings and adults, then

neighborhood seedling species identity would be relatively unimportant to each individual

seedling’s performance because seedlings would not compete intensely among themselves for

resources. On the other hand, if seedlings of relatively similar sizes compete intensely among

themselves, then neighborhood seedling species identity may be an additional, critical determi-

nant of seedling performance, and seedlings themselves may partition resources (Tilman 1982), as

many models of plant species coexistence assume (e.g., Pacala et al. 1996). The evidence avail-

able strongly suggests that competition is more intense between seedlings and adults than among

seedlings. The density of seedlings in closed-canopy forests is relatively low (Harms et al. 2004,

Moles and Westoby 2004b, Table 3.1). Observing the paucity of seedlings in the understory of

tropical forests, Wright (2002) conjectured that interactions among seedlings could be sufficiently

weak to allow the coexistence of every species able to tolerate the abiotic conditions of the under-

story, so long as frequency dependence kept the rarest from drifting to extinction.

No author, to our knowledge, has strongly promoted the idea that forest tree seedlings

frequently compete among themselves. Nevertheless, very few studies have directly assessed the

48

intensity of or potential for this interaction. High rates of frequency-dependent mortality befall

tree seedlings, increasing diversity (Webb and Peart 1999, Harms et al. 2000); this mortality is

consistent with, though not unequivocal evidence of, intense competition. Inter-seedling competi-

tion was intense in high-density masting populations of Acer saccharum seedlings (Taylor and

Aarssen 1989), but appeared weak in a study of tropical tree seedlings (Brown and Whitmore

1992). Conversely, much effort has been spent estimating the intensity of size-asymmetric compe-

tition. There is abundant evidence that shading by saplings (Marquis et al. 1986) and adults (e.g.,

Lewis and Tanner 2000) reduce the growth and survival of forest tree seedlings. Nevertheless,

determining the relationship between seedling population density and the intensity of inter-seed-

ling competition is essential to understanding the coexistence of tropical tree species.

In this study, we assess the intensity of, and potential for, competition among tree seed-

lings in two tropical rainforests. We test the relationship between seedling performance (growth

and survival) and population density by tracking seedling cohorts of three common tree species.

We predicted that intense competition among neighbors would generate an inverse relationship

between population density and seedling performance. The low density of tree seedlings we

observed in tropical rainforests initially motivated us to question the intensity of inter-seedling

competition (e.g., Harms et al. 2004). As competition is predicted to be most intense in high-den-

sity populations, we chose to study the densest patches of seedlings available in Panamanian and

Peruvian rainforests. Our study plots, which were dominated by one of three species, were more

than an order of magnitude denser than the mixed-species seedling layer typical of Neotropical

forests (95.0 ± 66.0 vs. 4.4 ± 4.5 individuals/m2 [weighted mean ± SD], Table 1). If competition is

found to be weak in such high-density plots, it is unlikely to be intense where seedling density is

lower, regardless of species composition.

49

Site Indiv./m2 (Mean ± SD)

N plots Reference

High-density plotsPouteria reticulata, BCI, Panama 168.7 ± 18.0 16 This studyBrosimum alicastrum, BCI, Panama 56.3 ± 10.0 13 This studyMatisia cordata, CCBS, Peru 52.7 ± 19.7 32 This study Mixed-species plots Paracou, French Guiana 18.5 ± 7.9 60 C. Baraloto, unpublished dataCCBS, Peru 6.4 ± 4.8 300 Harms et al. (2004)BCI, Panama 6.1 ± 1.5 300 Harms et al., (2004)Estación Biológica Los Amigos, Peru 6.0 ± 5.7 192 C. E. T. Paine, unpublished dataYasuni, Ecuador 6.0 ± 11 600 M. Metz, unpublished dataKm 41, Brazil 5.6 ± 2.1 300 Harms et al., (2004)Beni, Bolivia (occasionally hunted) 4.2 ± 0.1 229 Roldan and Simonetti (2001)Beni, Bolivia (intensively hunted) 3.9 ± 0.2 238 Roldan and Simonetti (2001)La Virgen de Sarapiquí, Costa Rica 2.6 ± 0.1 144 Capers et al., (2005)La Selva, Costa Rica 2.0 ± 1.0 120 Nicotra et al., (1999)Cay Rica, Costa Rica 1.5 ± 1.0 117 Nicotra et al., (1999)Chilamate, Costa Rica 1.4 ± 0.8 107 Nicotra et al., (1999)La Selva, Costa Rica 1.1 ± 0.7 300 Harms et al., (2004)

Table 3.1. Seedling densities in high-density experimental plots and in 11 Neotropical rainforests. Densities of woody dicots (plus juveniles of canopy palms) 10 – 50 cm tall are sorted by decreasing density. Study plots were significantly denser than mixed-species plots (ANOVA of weighted means: F2, 14 = 31.9, P < 0.0001). All sites receive at least 2200 mm of rainfall annually. Plot sizes ranged from 0.5 – 5 m2, but densities are scaled to 1 m2 for ease of comparison.

MATERIALS AND METHODS


We investigated intraspecific competition among seedlings at Barro Colorado Island,

Panama (BCI) and Cocha Cashu Biological Station, Peru (CCBS). Descriptions of the flora and

fauna of the two sites can be found in Croat (1978) and Gentry (1990), respectively. Annual

rainfall averages 2600 and 2200 mm at BCI and CCBS. At BCI, we performed density-reduction

experiments on plots dominated by Brosimum alicastrum Sw. (Moraceae) and Pouteria reticulata

50

(Engl.) Eyma (Sapotaceae). At CCBS, we performed similar manipulations on plots dominated by

Matisia cordata Bonpl. (Bombacaceae). All three species are widespread and moderately com-

mon, distributed from Panama through southeastern Peru. Henceforth, we refer to them by their

generic names. Pouteria and Matisia are canopy trees, whereas Brosimum is at times a canopy

emergent. All three species are moderately shade-tolerant; none are pioneer species (Croat 1978,

Foster and Janson 1985).

Experimental Reductions in Plot Density

Density-reduction experiments were performed beneath adult trees that had fruited within

the previous year. All seedlings were of similar sizes, and less than one year old at the initiation

of the density-reduction experiments. At BCI, we took advantage of an ongoing mammal-exclo-

sure experiment, which was established in 1993. We excluded mammals from a large plot (30

x 45 m) for five years. Very dense seedling patches, dominated by either Brosimum or Pouteria,

recruited naturally in the exclosure at extremely high densities. In 1997, we established 13 and 16

1-m2 plots in the high-density patches of Brosimum and Pouteria seedlings, respectively. For both

species, we randomly thinned the initially high-density plots to low, medium or high (unmanipu-

lated control) densities (15, 30, and 45-64 individuals/m2, respectively). For Pouteria, patches of

which were even more dense than those dominated by Brosimum, we established an additional

very-high-density treatment (148-188 seedlings/m2). Density treatments were applied randomly to

plots, which were separated by at least 1 m. Seedling densities in the BCI forest during the previ-

ous several years had been approximately 30 m-2, and our four treatment densities were designed

to represent half of the median forest seedling density, median seedling density, high seedling

density, and very high seedling density. Sample sizes for Brosimum plots were 4, 6 and 3 for low,

51

medium and high-density treatments, respectively; plots were evenly divided among treatments

for Pouteria (n = 4).

In 2004 at CCBS, we established three 0.5-m2 plots beneath each of eight Matisia adults,

separated by at least 250 m. We systematically selected and removed every second seedling to

increase nearest neighbor distances throughout each plot. Plots were thinned to low, medium, and

high (unmanipulated control) densities (12, 24, and 40 - 112 individuals/m2, respectively). Plots

beneath each adult Matisia were separated by at least 2 m, and were not protected from mammals.

Plots were evenly divided among treatments for Matisia (n = 8).

At both sites and for all three species, we clipped seedling above-ground biomass without

disturbing the soil, assuming that nutrient release from root decomposition was insignificant

(Coomes and Grubb 2000). All seedlings were independent of cotyledon reserves at the start of

the experiments, and no clipped seedlings of any species resprouted. Brosimum and Pouteria

seedlings were censused initially, and at 12 and 24 months, whereas Matisia was censused at 0,

8, 13 and 20 months. At each census, we measured the height of the apical meristem of all living

seedlings. In each census interval, we calculated the relative growth rate for individual height

(RGRht) as (ln(ht+1)-ln(ht))/T, where h is the individual’s height in cm at censuses t and t+1, and

T is elapsed time in days. Differences in individual RGRht were compared among density treat-

ments in a linear mixed model, blocked on plots. For Brosimum and Pouteria, survival rates were

compared among treatments at each census in a linear model with a binomial error distribution.

Because Matisia seedlings were censused three times, their survival functions could be estimated

using the Kaplan-Meier method. Matisia survival functions were compared with a log-rank test.

Survival analyses were performed only on seedlings present in the initial censuses.

52

Additionally, if competition were intense, the left-skewness of the distribution of plant

heights would be expected to increase through time (Obeid et al. 1967). We calculated the

skewness of the distribution of stem heights per treatment at each census period (pooling over

experimental replicates to achieve a reasonable sample size). All analyses were performed in R

2.3.1 (R Development Core Team 2006). All densities are reported per square meter for ease of

comparison.

The Overlap of Zones of Influence

The measurement of zones of influence (ZOIs) provides a powerful approach to distin-

guish between the often confounded processes of resource competition and consumer-mediated

(“apparent”) competition (Casper et al. 2003). Both resource-mediated and consumer-medi-

ated competition may generate density dependence in individual performance. But resource

competition additionally requires that the ZOIs of two plants overlap and that their growth be

resource-limited (Huston and DeAngelis 1994), whereas individuals in “apparent” competition

may be spatially isolated (Holt 1977). Plant growth is almost always resource-limited, except

following certain rare disturbances (Platt and Connell 2003). Therefore, the intensity of resource

competition can be accurately estimated by the extent of ZOI overlap. Following Huston and

DeAngelis (1994), we define a ZOI as the area within which an individual plant may affect the

availability of a limiting resource. Only if ZOIs overlap substantially may inter-seedling resource

competition affect seedling recruitment. All else being equal, increases in population density,

individual size, and small-scale clumping cause greater ZOI overlap and thus intensify resource

competition (Casper et al. 2003). Measurements of ZOIs may thus determine the potential for

resource competition among plants.

53

We quantified ZOIs in two types of plots at CCBS, separate from those of the density-

reduction experiment: high-density 0.5-m2 Matisia-dominated plots (n = 7), and randomly

located median-density 1-m2 mixed-species plots (n = 8). In Matisia-dominated plots, non-

Matisia seedlings (mean = 2.1 individuals per plot) were excluded from the analysis, whereas in

mixed-species plots, all vascular plants were included.

We estimated above- and below-ground zones of influence (ZOIs) independently. We

define an individual’s above-ground ZOI as the vertical projection of its leaf area. This definition

makes the simplifying assumption that mid-day sun is most important for plant growth in the

understory of closed-canopy tropical forests (Canham et al. 1990, Montgomery and Chazdon

2002). We report the overlap of seedling ZOIs, rather than the extent to which seedlings shade

each other, because the seedlings were of relatively uniform height (16.9 ± 3.8 cm [mean ± SD]),

and the relative height ranks of their leaves may be shuffled repeatedly over time.

We assessed above-ground ZOI overlaps photographically. A digital camera, equipped

with a wide-angle lens, was mounted on a tripod and centered above the plot (see Connell et

al. 1997, who developed the technique to study interactions among intertidal corals). The initial

image documented the exposed leaf area of each seedling. We immediately clipped and removed

the leaves not overlapped by any other leaf. Another image was taken of the leaves thus revealed.

The process was repeated up to six times until only bare stems remained. This final image

mapped the location of each individual’s point of contact with the ground. The above-ground ZOI

overlap of each seedling was calculated as the percentage of its leaf area overlapped by the leaves

of other seedlings. After we rectified the digital images in Adobe PhotoShop CS 8.0 (Adobe

Systems, Inc. 2003) to remove lens distortions, we compared successive images to determine

54

above-ground ZOIs and ZOI overlaps in ImageJ 1.36b (Rasband 2005). Measurements of above-

ground ZOIs by six people varied by less than 5%.

The extent of an individual’s below-ground ZOI depends upon the supply rate of limiting

resources, the rate at which the resources diffuse through soil, and the rate at which the individual

takes up the resources (Huston and DeAngelis 1994). An individual’s below-ground ZOI shrinks

with increased resource supply rate, and expands with increases in resource diffusion rates and

the individual’s resource uptake rate. If diffusion or uptake rates are slow, an individual’s ZOI will

be restricted to the immediate vicinity of its roots. Because the soils underlying CCBS are clayey

and relatively nutrient-rich (Osher and Buol 1998), we assume that resource supply rates are rela-

tively great, and diffusion is relatively slow. Given the difficulty of precisely determining spatial

relationships in the rhizosphere, we define a plant’s below-ground ZOI as a circle centered at its

stem with radius equal to the length of its longest lateral root (Casper and Jackson 1997, Casper et

al. 2003).

To measure below-ground ZOI overlap in the field, we gently loosened the soil around

each photographically mapped seedling, excavated it, and measured the extension of its lateral

roots. The overlap of below-ground ZOIs was calculated using the mapped locations of seedlings

and the radii of their ZOIs (see Figure 2 of Stohlgren 1993). An individual’s pairwise overlaps

were summed to calculate its overall below ground ZOI overlap.

Several potentially offsetting biases may have affected our estimates of below-ground

ZOIs. Small seedlings of tropical trees bear few lateral roots, meaning that they occupy highly

irregular polygons, the area of which is substantially smaller than our circular estimated ZOIs

(Casper and Jackson 1997). Moreover, root systems may vertically partition soil (Sala et al. 1989).

Because our ZOI model lacked a depth component, we may further overestimate ZOI overlaps.

55

On the other hand, our methods may have underestimated the radii of below-ground ZOIs as they

were based on the lengths of lateral roots, which may have broken during excavation (Coomes and

Grubb 2000). We believe our estimates of below-ground ZOI overlaps to be conservative, as we

likely overestimate the area of root systems. In estimating ZOI overlaps, we strove to minimize

the incidence of false negative results, the Type II error rate.

RESULTS

Over the 24-month experiment (20 months for Matisia), seedlings grew, on average, from

23 to 30 cm (30%), 21 to 26 cm (23%), and 17 to 24 cm (40%) for Brosimum, Pouteria, and Mati-

sia, respectively. Simultaneously, seedlings experienced a substantial risk of mortality: only 76,

58 and 15% of seedlings survived through the experiment, respectively.

Experimental Reductions in Plot Density

Inter-seedling competition did not explain the relationship between RGRht and plot den-

sity. Brosimum RGRht≠ was unrelated to plot density in either census period, or overall (F2, >290 <

2.8, P > 0.05; Fig. 3.1a). Pouteria RGRht did not differ among treatments after 12 months (F3, 880

= 1.6, P = 0.18), but did after 24 months and overall (F3, >669 < 4.3, P < 0.005; Fig. 3.1b). Surpris-

ingly, Pouteria seedlings grew more rapidly in higher density plots, contrary to the prediction

that competition among seedlings would reduce their growth in high density plots. This positive

density dependence suggests that these seedlings were in particularly favorable microsites or there

was intra-specific facilitation (Klironomos 2002). Matisia RGRht was unrelated to stem density

in any of the three census periods of the 20-month experiment, or overall (F2, >48 < 2.39, P > 0.10;

Fig. 3.1c). The significant decrease in Matisia heights between the 8 and 13-month censuses was

probably due to plants being damaged, but not killed, by falling debris. In sum, there was little

evidence that reductions in seedling density increased seedling growth rates for the three species.

56

Nor did inter-seedling competition

explain the relationship between seedling

survival and plot density. The survival of

Brosimum seedlings did not differ among

treatments in either census period or overall

(deviance < 0.57, df = 2, P > 0.75; Fig. 3.2a).

Survival of Pouteria seedlings, on the other

hand, differed significantly over the two-year

period (deviance = 14.00, df = 3, P = 0.0029;

Fig. 3.2b), though not in either 12-month

census interval (deviance < 6.53, df = 3, P

> 0.088). This difference in survival was

driven entirely by increased mortality in the

very-high-density treatment, in which initial

densities were an order of magnitude greater

than the average total density of seedlings

in Neotropical forests (Table 3.1). Based on

Kaplan-Meier analysis, Matisia survival func-

tions differed among treatments (χ2 = 124.7, P

Figure 3.1. Relative height growth rate (RGRht) of (a) Brosimum alicastrum, (b) Pouteria reticulata and (c) Matisia cordata seedlings grown in plots of experimentally reduced density. Bars to the right of the vertical dashed line indicate RGRht averaged over the course of the experiment. Bars indicate means ± 1 SEM.

57

< 0.0001, Fig. 3.2c), but the percent surviving

through 20 months did not (low = 0.06, 95%

CI = 0.02– 0.19, medium = 0.13, 95% CI =

0.07– 0.21, high = 0.17, 95% CI = 0.13– 0.23).

The differences in Matisia survival were driven

by initially greater survival in high-density

plots than in low- and medium-density plots – a

pattern that ran counter to the prediction that

competition among seedlings is intense. Overall,

the lack of relationship between density and

survival, except in the most extreme instance,

indicated that competition among seedlings of

the three species was weak.

The skewness of the distribution of

plant heights was examined for each density

treatment at each treatment period. Plant height

skewness varied among treatments and censuses,

but did not become more negative through time

in any species, regardless of treatment density.

Figure 3.2. Survival of (a) Brosimum alicastrum, (b) Pouteria reticulata and (c) Matisia cordata seedlings grown in plots of experimentally reduced density. Plot density reduced seedling survival only for Brosimum, and only in extremely dense plots. Points indicate means ± 1 SEM.

58

The Overlap of Zones of Influence

ZOI overlaps were analyzed in high-density Matisia-dominated plots and in ambient-den-

sity mixed-species plots. In Matisia-dominated plots, ZOI overlaps were minor, both above- and

below-ground, even though ZOI plots contained 41.4 ± 14.2 seedlings/m2 (mean ± SD). Seedlings

overlapped a median of just one other seedling above-ground (quartiles: 0 – 1), and experienced

two percent ZOI overlap (quartiles: 0 – 18%; Figure 3.3a). The leaves of approximately half of

the seedlings did not overlap at all with those of their neighbors. Below ground, seedlings experi-

enced a median of 13% ZOI overlap (quartiles: 0 – 38%). The root systems of 48% of individuals

did not overlap at all with their neighbors, and just 18% of seedlings overlapped more than 50%

with their neighbors (Figure 3.3b).

ZOI overlaps in mixed-species plots were even less than in Matisia-dominated plots

because of the overall lower density of seedlings (16.5 ± 7.0 individuals/m2, including ferns and

Figure 3.3. Histograms of above ground zone of influence (ZOI) overlaps in (A) high-density, Matisia-dominated plots and (B) typical mixed-species plots. Below-ground ZOI overlaps in (C) high-density, Matisia-dominated plots and (D) typical mixed-species plots. The ZOIs of most seedlings do not overlap those of their neighbors, neither above- nor below-ground, even in high-density plots. Bars represent the means of 7 and 8 plots ± 1 SEM for Matisia-dominated and mixed-species plots, respectively.

59

monocots). More than half (58%) of individuals did not overlap whatsoever with other individuals

above-ground (Figure 3.3c). The median overlap of seedlings was 0% (quartiles: 0 – 40%). Just

23% of seedlings overlapped more than 50% with their neighbors (Figure 3.3d). Altogether, there

was little potential for resource competition in high-density Matisia-dominated plots, and even

less in mixed-species plots of more typical density.

DISCUSSION

Our results suggest that competition among seedlings of three species of rainforest trees is

weak, even at densities an order of magnitude greater than those typical of seedlings in Neotropi-

cal forests (Table 3.1). Evidence of competition was not apparent in seedling RGRht or survival.

The distribution of plant heights did not become more left-skewed through time in high-density

plots, contrary to the predicted response of populations experiencing intense competition, and

unlike many experimental populations (especially populations of annual herbs, e.g., Obeid et al.

1967). The potential for inter-seedling resource competition was also low: substantial ZOI over-

laps were scarce in high-density Matisia-dominated plots, and even scarcer in ambient-density

mixed-species plots.

What were the proximate causes of weak competition in our experimental populations?

The paucity of substantial ZOI overlaps may have resulted from the relative richness of the soils

underlying EBCC (Osher and Buol 1998). In nutrient-starved habitats or high-light environments,

increased root:shoot allocation ratios may enlarge below-ground ZOIs (Coomes and Grubb 2000,

Montgomery and Chazdon 2002), which would increase ZOI overlaps unless offset by reduced

individual stature (Casper and Jackson 1997). Competition can affect above-ground seedling

architecture, sometimes even more than it affects growth and survival (Massey et al. 2006), but

that is unlikely to have occurred among seedlings of these three species, given their simple, un-

60

branched morphology. Our results – minor ZOI overlap, weak competition at high density, and

low densities of seedlings in Neotropical forests – suggest that resource competition among tree

seedlings may often be weak in the shaded understory of tropical forests.

What Killed the Seedlings in Our Study?

If competition among seedlings was not a significant source of mortality, what killed the

seedlings in our experiment? First, adult vs. seedling resource competition is intense and highly

asymmetrical (Coomes and Grubb 2000). Canopy trees can reduce understory light availability

to < 2% of the light available above the canopy (Canham et al. 1990, Montgomery and Chazdon

2002), which keeps seedling growth rates well below those found in canopy gaps. Trenching

experiments demonstrate that adults also suppress seedlings through below-ground competition

(Coomes and Grubb 2000, Lewis and Tanner 2000).

Second, a number of processes keep seedling densities low in forest understories (Wright

2002, Moles and Westoby 2004). These include consumers (seed predators, pathogens and herbi-

vores, Bell et al. 2006, Paine and Beck in press), disturbances in the form of debris falling from

the canopy (Clark and Clark 1989), and dispersal limitation (Svenning and Wright 2005). Brosi-

mum and Pouteria, protected from mammalian herbivores, had substantially lower mortality rates

than did Matisia, whose individuals were unprotected, suggesting that mammals killed some

seedlings. The fall of debris from the canopy was also probably an important source of mortality,

as many seedlings of all three species disappeared “without a trace”. Host-specific pathogens are

unlikely to have contributed to seedling mortality in this study, since they typically cause negative

density dependence, a pattern we did not observe. Dispersal limitation is also unlikely to have

reduced the intensity of competition in this study, as we located our plots beneath fruiting adults

61

specifically to maximize seedling density. If anything, limited dispersal increased the likelihood

of detecting negative density dependence.

Finally, competition among seedlings in the past may have reduced their density such that

no relationship between performance and density is observed currently. Though we cannot know

the intensity of competition prior to the initiation of our study, we infer that it had not occurred

among our seedlings prior to when the experiment began. In self-thinning stands, individuals may

initially be so small as to preclude interactions (Yoda et al. 1963). They crowd one another as they

grow, causing self-thinning, and intensifying competition. Unlike inter-trophic interactions, com-

petition rarely reduces population density so far as to preclude interactions among neighbors. For

our results to be explained by previous competition among seedlings, however, an additional pro-

cess would need to be invoked that so strongly reduced population density as to preclude current

competition. In sum, the absence of detectable present-day competition is unlikely to be attribut-

able to an ecological “ghost of competition past” (Connell 1980).

Interactions among seedlings may become intense when one or more of these factors is

weakened. For example, we observed negative density dependence in the survival of Pouteria

seedlings only where their initial density was ten times greater than the mean total density of

the seedling layer and they were protected by mammal exclosures (Figure 3.2b). Other situations

where competition among seedlings may intensify, particularly below ground, are in tree-fall gaps,

where survival and growth rates increase concomitantly with light availability (Schnitzer et al.

2005), or following community masting events, which can lead to extremely high seedling densi-

ties (Taylor and Aarssen 1989).

62

Competitive Irrelevance

The continuum of interactions proposed to occur among neighboring plants ranges from

strict competitive hierarchies, which result in resource partitioning (Tilman 1982, Chase and

Leibold 2003), to competitive equivalence, which underlies neutral models (Pacala 1988, Hubbell

2006). Our results suggest a possible complementary condition: competitive irrelevance.

Competitive irrelevance is a minimally interactive, unstable contributor towards coexis-

tence. Under competitive irrelevance, neighboring individuals are isolated by distance such that,

though the species may differ in competitive ability, interactions among individuals are reduced

in intensity. Isolation thus imposes an “equalizing force” (sensu Chesson 2000) on the competi-

tive ability of all individuals by reducing the intensity of competition between neighboring plants

to zero. Coexistence would be unstable in that, in the absence of interactions among individuals,

all mortality would be stochastic and density-independent, except for (potentially density-depen-

dent) predation. Competitive irrelevance would thus permit unstable coexistence without the

stringent and rarely realized requirement of strict competitive equivalence.

We emphasize that competitive irrelevance would apply only during a limited portion of

any life cycle, and even then, only to individuals of relatively uniform size. Moreover, competitive

irrelevance is most likely in habitats, such as the understory of closed-canopy forests, in which

the density and growth rates of individuals are diminished by factors other than competition,

as described above. Furthermore, competitive irrelevance makes no assumptions regarding the

potential asymmetry of interactions that may occur during other portions of the life cycle. As

plants grow and competition intensifies, individual performance would decline, avoiding the pre-

dictions of infinite population sizes that arise in non-interactive models (Caswell 1976).

63

Should competitive irrelevance hold for tree seedlings, the consequences for species

coexistence would be profound (Wright 2002). For example, the inferences that may be derived

from studies of seedlings conducted at extreme densities would be called into question (e.g., Tan-

ner et al. 2005). More generally, competitively superior and inferior species could coexist side

by side during the critical period of seedling recruitment, effectively isolated by distance from

competition. Nevertheless, the temporal duration of the phase of competitive irrelevance remains

to be investigated. Competitive hierarchies cannot give rise to competitive exclusion if potential

competitors do not interact. Competitive irrelevance could thus permit numerous individuals to

establish. If more species can establish or even reach reproductive maturity without competing

among individuals of the same size class, then long-term coexistence would be facilitated. In

other words, if competition among seedlings were absent or weak, other processes, such as neutral

dynamics, size-asymmetric competitive hierarchies, or interactions with consumers, would by

necessity prevail as those structuring the forest understory.

64

CHAPTER 4 — SEED PREDATION BY NEOTROPICAL RAINFOREST MAMMALS INCREASES DIVERSITY IN

SEEDLING RECRUITMENT*

INTRODUCTION

The vast majority of seeds that fall in forest understories fail to recruit as seedlings. Mor-

tality rates during seedling recruitment – the seed to seedling transition – are greater than at any

other life stage (Muller-Landau et al. 2004). This mortality tends to be negatively density depen-

dent, which, in turn, increases diversity of the seedling layer (Webb and Peart 1999, Harms et al.

2000, Hille Ris Lambers et al. 2002). The filters that limit seedling recruitment thus contribute to

species coexistence. It is therefore essential to study the causes of seed and seedling mortality to

understand the processes that maintain forest diversity. However, the agents of mortality are rarely

identified, because the death of an individual seed or seedling is unlikely to be observed through

sampling. For these reasons, our understanding of the mechanisms that contribute to seed and

seedling mortality in forests remains incomplete.

Terrestrial mammals affect all stages of seedling recruitment. Many terrestrial mammals

are predators of seeds and seedlings, which can dramatically affect patterns of seedling recruit-

ment (Terborgh et al. 1993, Ostfeld et al. 1997, Notman and Gorchov 2001, Silman et al. 2003,

DeMattia et al. 2004). Terrestrial mammals also disperse seeds, (e.g., Brewer 2001) and trample

seedlings (Clark and Clark 1989). Nevertheless, following Beckage and Clark (2005), we refer

to terrestrial mammals as predators of seeds and seedlings, understanding that their effects on

seedling recruitment encompass various processes. Mammals consume more seeds than do inver-

* Reprinted here by permission of Ecology. Citation: Paine, C. E. T. and Beck, H. In Press. Seed predation by neotropical rainforest mammals increases diversity in seedling recruitment. Ecology.

65

tebrates (Holl and Lulow 1997, Notman and Gorchov 2001), and small mammals may consume

more seeds than large mammals (DeMattia et al. 2004). Excluding terrestrial mammals can

increase seedling recruitment and survival (Ostfeld et al. 1997, Connell et al. 2005), change seed-

ling community composition (DeMattia et al. 2006) and increase seedling growth rates (Wahungu

et al. 2002). Nevertheless, many studies of mammalian predation on seeds and seedlings have

been of short duration (≤ 6 months; Holl and Lulow 1997, Notman and Gorchov 2001, Wahungu

et al. 2002), or focused on few species (≤ 3 species; Asquith et al. 1997, Connell et al. 2005,

Norghauer et al. 2006), limiting their scope for inference.

Among trees, two axes of variation that may interact with mammalian predation to affect

seedling recruitment are population density and seed mass. An inverse relationship between

population density and the recruitment, survival or growth of conspecifics constitutes negative

frequency dependence, which is common in tropical and temperate forests (Webb and Peart 1999,

Harms et al. 2000, Hille Ris Lambers et al. 2002, Peters 2003, Uriarte et al. 2005). Recruitment

rates that are greater for rare species than for common species would indicate a rare-species

advantage, the critical factor to generate frequency dependence (Connell 1978). We use “fre-

quency dependent” to indicate a relationship with the abundance of conspecific adult trees, rather

than with seed or seedling density. Studies focused on one or few species indicate that predator

aggregation, rather than resource competition, is likely to be the primary mechanism causing fre-

quency dependence (Augspurger 1984, Blundell and Peart 1998). But the processes that generate

frequency dependence have rarely been elucidated at the community level.

The per-capita likelihood of seedling recruitment increases with increasing seed size

(Moles and Westoby 2004), but with their greater fecundity, seeds of small-seeded species may

arrive at a more diverse set of microsites than those of large-seeded species. Large-seededness

66

reduces susceptibility to biotic and abiotic hazards like deep shade or leaf litter (Molofsky and

Augspurger 1992, Saverimuttu and Westoby 1996). But large-seeded species may be more suscep-

tible to predation because, all else being equal, foragers can maximize their energetic return by

preferentially consuming large food items (Schoener 1971, Brewer 2001, Moles et al. 2003). The

advantages that accrue to large-seeded species during seedling recruitment are strong, but may

dissipate rapidly if predation is intense (Moles and Westoby 2004a).

Objectives

Our primary objective was to determine how terrestrial mammalian predation, adult

abundance, seed mass, and their interactions affect seedling recruitment and contribute to species

coexistence. We tested their effects on all stages of seedling recruitment: the survival of seeds,

the survival and growth of seedlings, and the density and diversity of recruited seedlings.

First, we determined whether groups of terrestrial mammals differ in their predation rates.

In particular, we asked whether the biomass or abundance of each of three size-classes of mam-

mals better explains the variance in each stage of seedling recruitment. Second, we determined

the effects of adult abundance and seed mass on each stage of seedling recruitment. Third, we

determined the strength of the interactions between mammalian predation and seed mass, and

predation and adult abundance during each stage of seedling recruitment. Our overall goal was to

understand how predation during seedling recruitment affects the diversity of the seedling layer.

METHODS


This study was conducted at Estación Biológica Cocha Cashu (EBCC) in Peru’s two-mil-

lion ha Manu National Park (~12° S, 71° W, ~350 m elevation, see site descriptions in Terborgh

1983, Terborgh 1990). The forest-covered floodplain of the Manu River is extremely diverse, with

67

almost 350 species of trees that attain a diameter of 10 cm at breast height (Foster 1990). The site

is characterized as Tropical Moist Forest (Holdridge 1947). Average annual precipitation is 2200

mm, falling mainly between October and April.

EBCC is one of few sites worldwide where diverse communities of terrestrial mammals

remain intact and accessible for study. The vastness and physical isolation of Manu National

Park has facilitated the preservation of EBCC’s mammal community (Terborgh 1999). It is thus

an ideal location to detail the varied effects of terrestrial mammals on seeds and seedlings. We

divide the terrestrial mammal community into three size classes. Small mammals weigh less

than 1 kg, comprising mice (Muridae) and spiny rats (Echimyidae). Medium-sized mammals are

caviomorph rodents and include green acouchis, pacas, and agoutis (Myoprocta spp., Agouti paca,

and Dasyprocta variegata, respectively), which weigh 1 to 12 kg. Large mammals weigh more

than 20 kg, and are predominantly peccaries (Pecari tajacu and Tayassu pecari, Tayassuidae), but

also include deer (mostly Mazama spp, Cervidae) and tapirs (Tapirus terrestris, Tapiridae). Tapirs

and both species of peccaries that occur at EBCC are threatened with extinction, listed in CITES

Appendix II. Seeds and fruit form an important component of the diet of all three size-classes

(Beck-King et al. 1999, Vieira et al. 2003, Beck 2005).

Seeds were included in the exclosure experiment based upon three criteria. First, fruit had

to be single-seeded and their seeds had to be sufficiently large to be easily cleaned, sown, and

monitored in the field. Second, fruit needed to be available during the two periods of placement

of seeds into the exclosures, (I) April – June 2004 and (II) February 2005. And third, fruit-

ing adults had to be sufficiently common and fecund to provide 400 seeds for placement in the

exclosures. These criteria yielded 14 species, representing 12 families, all of which are primarily

mammal-dispersed (Foster and Janson 1985) including eleven trees, two palms, and one liana

68

(Sparattanthelium tarapotanum [Hernandiaceae]), which remains free-standing until reaching at

least 3 m in height (C. E. T. Paine, personal observation). All species reach the canopy as adults.

Adult abundance was determined in three permanent plots totaling 10.25 ha in the EBCC trail

system (data provided by J. Terborgh, Duke University). These plots were representative of mature

floodplain forest, and were interspersed with the locations of exclosures. Adult abundance of the

14 focal species ranged from 0.08 – 37 trees/ha (median = 0.94 trees/ha), encompassing the entire

range of adult abundances observed among canopy trees in this forest. Our use of the descrip-

tors ‘common’ and ‘rare’ refer only to local conspecific abundance of adults, rather than to seed

or seedling density. Seed mass of the 14 species ranged from 0.05 – 3.9 g (median = 1.8 g). Adult

abundance and seed mass were not significantly correlated among the 14 species used in this

experiment (R2 = 0.09, P = 0.31). Species names and distinguishing characteristics are presented

in Table 4.1.

Experimental Design

To determine the individual effects of small, medium, and large mammals, we used exclo-

sures that differed only in their permeability to each size-class, which were designed and built by

Beck and Terborgh (in prep.). One year after that study was completed, we reinforced the existing

exclosures for the current study. The 40 exclosures were established in eight randomly located

blocks in an area of 3 km2 of mature lowland rainforest at EBCC. The mean distance among

blocks was 1.1 km (minimum distance between blocks: 250 m). Within each block, we located

five 2 x 2-m exclosure cages, one of each of five types, twenty meters apart along a randomly ori-

ented transect.

The five types of exclosures were designed to preclude or permit the entry of small,

medium or large mammals (Figure 4.1). The name of each exclosure indicates the size-class of

69

mammals that was permitted access. NONE exclosures were 90-cm tall wire hardware cloth

(mesh size 1 cm), reinforced with 8-mm diameter rebar at the corners and the middle of each

side. No terrestrial mammals could enter NONE exclosures. SMALL exclosures were identical

to NONE, with the exception that 7 x 7 cm holes were cut in the bottom edge of the mesh walls.

SMALL exclosures permitted the entry of small mammals, while precluding the entry of medium

and large mammals. MEDIUM-LARGE exclosures consisted of 20-cm tall walls of sheet metal

flashing that small mammals were unable to climb over. Medium and large mammals could, in

Family Species Period Adult density (ha-1)

Seed mass (grams)

Annonaceae Duguetia quitarensis Benth. I 3.60 0.4 ± 0.07 (9)Arecaceae Iriartea deltoidea Ruiz & Pav. I 32.30 3.9 ± 0.4 (10)

Socratea exorrhiza (Mart.) H. Wendl.

I 0.08 3.4 ± 0.6 (10)

Clusiaceae Calophyllum brasiliense Cambess.

I 0.38 2.5 ± 0.8 (44)

Combretaceae Buchenavia grandis Ducke I 0.17 1.9 ± 0.5 (20)Ebenaceae Diospyros pavonii (A. DC.) J.F.

Macbr.I 0.17 0.6 ± 0.1 (84)

Diospyros subrotata Hiern. I 2.33 0.8 ± 0.1 (37)Flacourtiaceae Casearia sp. nov. II 0.17 0.6 ± 0.1 (25)Hernandiaceae Sparattanthelium tarapotanum

Meisn.II 0.03 0.5 ± 0.1 (26)

Moraceae Clarisia racemosa Ruiz & Pav. II 3.80 1.8 ± 0.9 (25)Myristacaceae Otoba parvifolia (Markgr.) A.H.

GentryII 37.30 1.8 ± 0.2 (25)

Olacaceae Heisteria nitida Engl. I 0.25 0.2 ± 0.04 (5)Rubiaceae Genipa americana L. II 0.94 0.05 ± 0.004 (25)Verbenaceae Vitex cymosa Bertero ex Spreng. II 0.40 0.5 ± 0.2 (25)

Table 4.1. Species sown in mammal exclosures, sorted by family. All species are canopy trees, except Sparattanthelium tarapotanum, a canopy liana. Seeds were sown in two periods, in 2004 and 2005. Seed masses are reported ± SD, with sample size in parentheses.

Note: Sparattanthelium tarapotanum, a canopy liana, is not censused in the EBCC tree plots. Extensive cruising has shown it to be very rare (C. E. T. Paine, personal observation).

70

� � ��

� � ��

��

��

��

� � ��

� � ��

� � ��

��

Figure 4.1. Schematic of the five different types of mammal exclosures. Checkmarks indicate the mammal size-classes that could enter each exclosure type. See text for detailed descriptions.

contrast, step over the flashing to enter the exclosure. MEDIUM combined the sheet metal of

MEDIUM-LARGE with a wrapping of barbed wire between 45 and 90 cm high. The sheet metal

barred the entry of small mammals, and the barbed wire barred the entry of large mammals.

Only medium-sized mammals had access to MEDIUM exclosures. Finally, ALL treatments were

marked with rebar at the four corners, permitting the entry of mammals of all three size-classes.

The walls of all exclosures (except ALL) were dug 5 cm into the soil to prevent the entry of

burrowing mammals. No evidence of burrowing mammals was detected in the duration of the

experiment. All exclosures had open tops, which facilitated monitoring and allowed the entry

of granivorous birds, such as wood-quail (Odontophorus) and cracids (Cracidae), and arboreal

mammalian granivores, such as squirrels (Sciuridae). All exclosures were likewise accessible to

invertebrates and microbial pathogens.

We prepared seeds for placement as fruit matured. We collected freshly fallen fruit from

at least five adult trees located throughout the 12-km2 area of the EBCC trail system, thereby

controlling for genetic and environmental factors. Fruit pulp was removed manually by methods

that varied among species. As necessary, seeds were washed or scrubbed to remove pulp, then

71

air-dried. Seeds that bore any sign of insect or fungal damage, such as exit holes or visible hyphae,

were discarded. Seeds of non-floating species were subjected to a flotation test: any floating seed

was discarded. Seeds were stored in cotton bags in ambient humidity for fewer than seven days

prior to placement.

We placed seeds in exclosures as follows. Conspecific seeds were laid in groups of ten on

top of the leaf litter in each exclosure (400 seeds/species in total). The density and clumping of

seeds we placed into exclosures thus mimicked those naturally dispersed by frugivorous monkeys

(Russo 2005), though our seed groups were placed without feces. Groups were placed in the

corners and the middle of each side of each exclosure at least 25 cm from the exclosure walls, so

that mammals outside the exclosure could not disturb the seeds. The conspecific groupings were

unlikely to affect seed removal rates (Notman et al. 1996), and facilitated repeated monitoring.

The locations of conspecific seed groupings were marked with colored plastic toothpicks, but the

seeds themselves were not individually marked.

Seeds, and as they germinated, seedlings, were monitored repeatedly. Eight species were

monitored for 790 days (12 censuses; Period I), whereas the remaining 6 species were monitored

for 509 days (six censuses; Period II). Intervals between censuses lengthened over the course of

the experiment from an initial interval of 2 weeks to a final interval of 6 months. At each census,

all remaining seeds were scored as alive, germinated, or missing. We use seed removal as a proxy

for mortality given the uncertainty in the precise fate of missing seeds (Vander Wall et al. 2005).

We thus scored missing seeds as dead, whereas intact or slightly damaged seeds were scored as

alive. Seeds that were gravely damaged, e.g., engulfed by fungal hyphae, were scored as dead. At

each census, we searched for missing seeds, and maintained the exclosures to ensure their ongo-

ing effectiveness. Leaf litter disturbed during a census was replaced. Because seeds were placed

72

in small groups, it was easy to distinguish and exclude naturally dispersed seeds. Finally, at each

census, we marked seedlings individually as they germinated, and measured the height of their

apical meristems.

Data Analysis

Seed survival was estimated in two ways: the total number of seedlings observed to ger-

minate per 10-seed grouping, and the median time to seed removal. The total number of seedlings

observed provides a minimal estimate of the fraction of placed seeds that survived to germination.

Because seeds were placed in exclosures in groups of 10, the time to median seed removal can

be interpreted as an estimate of the survival time of an average seed. For most 10-seed groups

(557 of 560), no seeds remained viable at the end of the experiment, and the time to median seed

removal could be calculated empirically (non-parametrically). In each of the remaining three

groups, one seed remained alive at the termination of the experiment. In these cases, Weibull sur-

vival functions were used to estimate median survival times (Klein and Moeschberger 1997).

We evaluated seedling performance with two response variables: survival and relative

growth rate. Post-germination seedling survival was calculated as the density of recruited seed-

lings divided by the total number of seedlings observed to germinate. We defined a seedling to

have recruited if it lived 509 days after sowing (the maximum time observed on both Period I and

Period II species). Mean height relative growth rate (RGRht) was calculated for each individual

seedling observed alive more than once over the course of the experiment, excluding those (n = 3

individuals) crushed under falling debris. RGRht was calculated as (ln(ht+1)-ln(ht))/T, where h is the

individual’s height in cm at times t and t+1, and T is time in days (Hoffmann and Poorter 2002).

Individual RGRht was averaged over each conspecific seed group. Seedling performance data

73

were unbalanced because of the differences in seedling germination across species and treatments,

but the statistical models used were robust to unbalanced data (Littell et al. 2002).

Finally, we evaluated the density and diversity of recruited seedlings. Density and spe-

cies richness of recruited seedlings was taken as the number of seedlings and species in each

exclosure at 509 days. Exclosures in which many seedlings recruited were speciose, resulting in a

significant positive relationship between the density and species richness of recruited seedlings (r

= 0.93, F1,38 = 274.2, p < 0.0001). We therefore use species richness per stem as a metric of diver-

sity because it controls for differences in stem density among exclosures (Hubbell et al. 1999).

Our true interests in this analysis regarded the effects of each size-class of mammals as

predators, rather than overall treatment effects. We used a priori, one degree of freedom orthogo-

nal contrasts to test the separate effects of small, medium, and large mammals on each stage of

seedling recruitment. The impact of each mammal size-class was determined by contrasting each

response variable between exclosure treatments that differed only in their permeability to that

size-class. Accordingly, we contrasted NONE with SMALL exclosures to estimate the effect of

small mammals, NONE – MEDIUM for medium mammals and MEDIUM – MEDIUM-LARGE

for large mammals. Finally, we used linear mixed models to determine the effects of adult

abundance and seed mass, and their interactions with each mammal size class, on each stage of

seedling recruitment.

We evaluated the relationships between each aspect of seedling recruitment and treat-

ments, species, and blocks in a split-plot design. Within blocks, treatment was the main-plot effect,

and species the sub-plot effect. An examination of residual plots revealed heteroscedasticity in

time to median seed removal and species richness per stem, which was adequately resolved with

a log transformation. Other variables conformed to the assumption of homogeneous variance.

74

Transformed variables were back-transformed for plotting. Analyses were performed in SAS 9.1

(SAS Institute, Cary, NC 2005).

RESULTS

Mammalian predators strongly reduced seed and seedling survival. The number of seeds

surviving plus seedlings germinated was consistently greater in exclosures from which all terres-

trial mammals were excluded (NONE) than in exclosures open to all mammals (ALL; Figure 4.2).

Survival rates in each treatment were rank-concordant between periods, indicating the pattern of

seed survival among exclosures to be robust to changes in species composition. As an example

using Period I species, as of November 2004, approximately 200 days after sowing, the number of

seeds plus seedlings in NONE exclosures was 18 times greater than in ALL exclosures. Mortality

was substantial even in NONE exclosures (45% through 200 days), indicating that many seeds

succumb to the combined attacks of pathogens, invertebrates, arboreal mammals, and granivorous

birds. Nevertheless, survival differed widely among exclosure treatments, indicating that mamma-

lian predation explained a substantial fraction of seed and seedling mortality.

Exclosure treatment, seed species, and their interaction had significant effects on time to

median seed removal, seed survival, and recruited seedling density (Table 4.2). Seedling survival

differed significantly among treatments and species, whereas RGRht differed only among spe-

cies. Data on seedling growth and survival were unbalanced enough as to preclude estimation

of treatment by species interactions. On a per-stem basis, i.e., after normalizing for stem density,

treatments differed significantly in the species richness of recruited seedlings.

Mammal size-classes differed in the strength of their effects on seed survival and

seedling performance (Figure 4.3). Large mammals had no significant effect on any stage of

seedling recruitment. In contrast, the exclusion of small and medium-sized mammals significantly

75

Exclosure Treatment Species InteractionResponse variable df F P df F P df F PTime to median seed removal (days)

4, 28

15.2 < 0.0001 13, 455

52.2 < 0.0001 52, 455

1.7 0.0034

Seed survival (percent)

4, 28

20.6 < 0.0001 13, 455

10.5 < 0.0001 52, 455

3.6 < 0.0001

Seedling survival (percent)

4, 28

8.6 0.0001 13, 507

8.2 < 0.0001 - - -

Seedling RGR (cm-1 cm-1 day)

4, 128

1.1 NS 12, 128

3.0 0.0010 - - -

Seedling recruitment (individuals)

4, 28

17.3 < 0.0001 13, 455

8.1 < 0.0001 52, 455

3.1 < 0.0001

Species richness per stem

4, 23

7.5 0.0005 - - - - - -

Table 4.2. Overall treatment effects of mammal exclosures, seed species, and their interactions. Data for seedling survival and relative growth rate were so unbalanced as to preclude estimation of treatment by species interactions. NS = not significant.

Figure 4.2. The survival of seeds and resultant seedlings differed widely among the five types of mammal exclosures. Seeds were added in two periods, in 2004 and 2005. For clarity, only Period I seeds are shown. Two species (Diospyros pavonii and Duguetia quitarensis) were sown in June 2004, temporarily increasing the number of seeds and seedlings.

0

2

4

6

8

10

Apr 04 Oct 04 Apr 05 Oct 05 Apr 06Date

NONEMEDIUM-LARGEMEDIUMSMALLALL

76

increased the time to median seed removal, seed survival and post-germination seedling survival.

On average, excluding small and medium mammals doubled the time to median seed removal, tri-

pled seed survival, and doubled seedling survival. The joint effect of excluding small and medium

mammals was to triple the density of recruited seedlings. Seedling RGRht, on the other hand, was

unaffected by any mammal size-class. Furthermore, predation by small mammals significantly

increased the per-stem species richness of recruited seedlings. Overall, the effects of small and

medium mammals were significantly greater than those of large mammals on all aspects of seed-

ling recruitment.

We then determined the relationship between adult abundance and each stage of seedling

recruitment. Seeds of species that are rare as adults survived significantly longer as seeds than

did relatively common species (F = 4.78, P = 0.029), but no other stage of recruitment differed

between rare and common species (F ≤ 1.90, P > 0.16, data not shown). On the other hand,

species differing in adult abundance fared very differently when interacting with mammalian

predators (Figure 4.4). Small mammals rapidly removed seeds of common species, and strongly

reduced the relative abundance of their seedlings. Though they did not disproportionately affect

seed removal rates or survival, medium-sized mammals similarly reduced the relative abundance

of recruited seedlings of common species. A 10-fold increase in adult abundance predicted 17%

greater seedling recruitment in the absence of small or medium mammals. Large mammals, hav-

ing had no effect on any stage of seedling recruitment, did not interact significantly with adult

abundance (F ≤ 1.15, P > 0.29). Together, these results suggest that small and medium-sized mam-

mals may impose frequency-dependent mortality by disproportionately preying upon the seeds of

common species, generating a rare-species advantage.

77

A

B

C

D

E

F

0

40

80

120DaystoMedian

Seed

Removal

ExcludedAllowed

t = 4.8P < 0.0001

t = 2.5P = 0.021

t = 0.23NS

0

20

40

t = 7.1P < 0.0001

t = 5.6P < 0.0001

t = 0.46NS

0

20

40

Seedling

Survival(%)

t = 3.6P = 0.0014

t = 3.0P = 0.0053

t = 0.53NS

-0.001

0.000

0.001

0.002t = 0.40NS

t = 0.80NS

t = 1.46NS

0

1

2

3

SeedlingRecruitment

(individuals)

t = 6.2P < 0.0001

t = 5.1P < 0.0001

t = 0.32NS

0.0

0.5

1.0

Small Medium Large

SpeciesRichness

perStem

t = 3.1P = 0.0048

t = 1.72NS

t = 0.01NS

RGR h

t(cm-1cm

-1day)

Mammal Size-Class

Seed

Survival(%)

Figure 4.3. Aspects of seedling recruitment in exclosures open or closed to each group of mammals. (a) Time to median seed removal, (b) mean seed survival, (c) seedling survival, (d) height relative growth rate, (e) density of recruited seedlings, and (f) species richness of recruited seedlings in exclosures which the specified mammal size-classes were allowed to enter, or excluded. Each bar represents the mean of 112 values. Error bars indicate 95% confidence intervals.

78

Seed mass had limited direct effects on seedling recruitment, as it was correlated only

with seed removal rates. Smaller-seeded species were removed more rapidly than larger-seeded

species (F = 28.12, P < 0.0001). Large and small-seeded species did not differ in any other stage

of seedling recruitment (F < 2.05, P > 0.15, data not shown). When interacting with mammalian

predators, on the other hand, any advantage accruing to large-seeded species disappeared (Figure

4.5). Small and medium mammals disproportionately reduced the survival of large-seeded spe-

cies, and reduced the relative abundance of their seedlings. A 1-g increase in seed mass predicted

a 10% increase in seedling recruitment in the absence of small mammals. Large mammals, hav-

ing had no effects on any stage of seedling recruitment, did not interact significantly with seed

mass (F ≤ 0.49, P > 0.49). Although small and medium-sized mammals reduced the overall

recruitment of seedlings, they both increased the relative abundance of seedlings germinated

from small seeds.

Summary

Surprisingly, large mammals had no effect on any aspect of seedling recruitment. Rather,

smaller mammals affected all stages of seedling recruitment more strongly than did large mam-

mals. Small and medium mammals significantly reduced the survival of seeds and seedlings,

which jointly reduced seedling density. Small mammals increased the per-stem species richness

of recruited seedlings by disproportionately preying upon seeds of common species and large-

seeded species, which increased the relative abundance of seedlings of rare and small-seeded

species. Altogether, predation by small mammals significantly increased the diversity of the seed-

ling layer.

79

Figure 4.4. Interacting effects of adult abundance and small- and medium-mammal access on seed survival and seedling recruitment. Predation by small and medium mammals both interact significantly with adult abundance. Where mammals were excluded, survival and recruitment increase with adult abundance (X’s and dashed lines). But in exclosures to which they had access, small and medium mammals disproportionately consumed common species, significantly reducing their recruitment (circles and solid lines). Statistics on each panel indicate the significance of the interaction of adult abundance and mammal access. Large mammals had no overall or interaction effects on any response variable, and are not plotted.

0

100

200 P = 0.0356 NS

0.0

0.4

0.8NS NS

0.01 0.1 1.0 10 0.01 0.1 1.0 10

0.0

0.4

0.8P = 0.0004 P = 0.0056

Adult Abundance (ha-1)

SeedlingSurvival

Seed

Survival

DaystoMedian

Seed

Rem

oval

Small Mammals Medium Mammals

Figure 4.

80

0

100

200 NS NS

0.0

0.4

0.8P < 0.0001 P < 0.0001

0 1 2 3 4

0.0

0.4

0.8P = 0.0264

0 1 2 3 4

P = 0.0363

Seed mass (g)

Small Mammals Medium MammalsSeedlingSurvival

Seed

Survival

DaystoMedian

Seed

Rem

oval

Figure 5.

Figure 4.5. Interacting effects of seed mass and small- and medium- mammal access on seed survival and seedling recruitment. Predation by small and medium mammals both interact significantly with seed mass: small and medium mammals disproportionately consume large-seeded species, significantly reducing their recruitment. Symbols and lines are as in Figure 4.4. Statistics on each panel indicate the significance of the interaction of seed mass and mammal access. Large mammals had no overall or interaction effects on any response variable, and are not plotted.

81

DISCUSSION

Mammal Effects

Both large and small mammals may be expected to have substantial effects on seedling

recruitment. In intact Neotropical forests, the biomass of large terrestrial mammals exceeds that

of medium or small mammals (400, 57, and 14 kg/km2, respectively; Janson and Emmons 1990).

On the other hand, small mammals are by far the most abundant size-class of mammals in Neo-

tropical forests. At EBCC, large, medium and small mammals occur at densities of 12, 14 and 410

individuals/km2, respectively (Janson and Emmons 1990).

Previous experimental and comparative studies have suggested that small mammals are

more important predators of seeds and seedlings than are large mammals (DeMattia et al. 2004,

Asquith and Mejia-Chang 2005). A study conducted at EBCC from 1988 to 1990 indicated that

small mammals were more important seed predators than large mammals (Terborgh et al. 1993).

However, the abundance of peccaries at EBCC was greatly depressed at that time, perhaps by an

epidemic disease (Silman et al. 2003). We expected that large mammals would play a stronger

role in our study because their density, especially that of peccaries, was greater at EBCC at the

time of our study than at any site where exclosure studies have previously been conducted. Instead,

small mammals were the dominant predators of seeds and seedlings, confirming the results of

previous investigations.

Why did small mammals affect seedling recruitment more than large mammals? Small

mammals are ubiquitous in the forest understory, with small, sometimes non-overlapping territo-

ries (Emmons 1982, Beck et al. 2004). Conversely, large mammals are scarce and occupy large

home ranges, as exemplified by white-lipped peccaries (Tayassu pecari). White-lipped pecca-

ries range through the understory of intact rainforests in “nomadic” herds of 50 – 200 individuals.

82

As they “bulldoze” through the understory of Neotropical forests, their effects are intense, but

tightly localized (Beck 2006). Therefore, the interval between successive visits of a peccary herd

to any point in the understory is likely to be on the order of months to years. For small rodents,

on the other hand, the recurrence interval is on the order of hours to days. Any single seed in

the understory is thus more likely to be located by a rat than a peccary. The difference in effects

between small and large mammals is essentially a sampling issue: in both space and time, small

mammals sample the forest understory more thoroughly than do large mammals. Moreover, small

mammals can consume even hard palm seeds (Brewer 2001, Beck and Terborgh 2002). None of

our study species had such strong mechanical defenses as to be invulnerable to small mammal

attack (cf. Terborgh et al. 1993). Small mammals thus more strongly affected seedling recruit-

ment than did large mammals.

In addition to their intense predation on seeds, small and medium mammals imposed

stronger herbivory on seedlings than did large mammals. We found that only smaller mammals,

and not large mammals, significantly reduced seedling survival. The only other study to partition

seedling herbivory between size-classes of tropical mammals gave largely inconclusive results

(DeMattia et al. 2006). Generally, deer and tapirs are considered the most important terrestrial

mammalian herbivores in Neotropical forests (Bodmer and Ward 2006). But they are relatively

scarce near EBCC, possibly because of the abundance and diversity of large carnivores at the

site. Generalist phytophagous insects, such as leaf-cutter ants (Atta spp.), can be very important

herbivores in tropical forests (Terborgh et al. 2001). Because they had free access to all exclosures,

however, their effects were not evaluated in this study. Ours is thus the only study to have identi-

fied rodents as the dominant mammalian herbivores of seedlings in a tropical forest.

83

Herbivory by small mammals reduced seedling survival, but no size-class of mammals

significantly affected relative height growth rates. Mammals evidently killed seedlings outright,

rather than merely browsing them, just as Beckage and Clark (2005) found in North Carolina.

In contrast, two Australian studies found that seedlings in exclosures grew more rapidly than

un-protected seedlings (Osunkoya et al. 1993, Wahungu et al. 2002). This is an important result,

as it indicates that mammalian herbivores may remove seedlings from the rainforest understory.

They thus generate open space that is subsequently available for colonization, rather than simply

slowing seedling growth rates. Should this open space be colonized by previously absent species,

small-scale species richness would increase.

Predation Increases Species Diversity

Small mammals increased species diversity, measured as per-stem species richness.

Though they reduced the species richness of recruited seedlings, small mammals reduced

seedling density even more sharply. Thus, they increased species richness on a per-stem basis.

Moreover, this effect was non-random with respect to the 14 species we added as seeds. Specifi-

cally, small mammals disproportionately preyed upon common and large-seeded species. Thus,

the increase in per-stem species richness was generated by increases in the relative abundance of

rare and small-seeded species.

Adult abundance and seed mass were weak predictors of success in seedling recruit-

ment overall. Rare species and large-seeded species fared better than common and small-seeded

species only initially, in terms of the time until median seed removal. Thereafter, neither adult

abundance nor seed mass predicted seedling performance or recruit density. The result for seed

size is consistent with the analysis of Moles and Westoby (2004) that the advantages of large seed

size dissipate rapidly.

84

Nevertheless, adult abundance and seed mass each interacted strongly with mammalian

predation as small and medium mammals preferentially consumed both common and large-

seeded species. In exclosures to which they had access, small and medium-sized mammals

reduced the recruitment of large-seeded species to levels comparable with those of small-seeded

species. Predation thus increased the relative abundance of small-seeded species in the seedling

layer. Insofar as seed size may correlate with competitive ability of young seedlings (Saverimuttu

and Westoby 1996), predation by small and medium mammals increased the relative abundance

of competitively inferior species. Similarly, small and medium mammals disproportionately con-

sumed seeds of common species, increasing the relative abundance of rare species. Why rodents

may preferentially prey upon seeds of common species is unknown, though they may develop

stronger search images for relatively common species, which may be more consistently rewarding

than rare species.

Adult abundance and seed mass captured some of the variation among species in this

experiment, but a substantial portion remained unexplained. Species showed idiosyncratic

responses to mammalian predation, shaped by details of their morphology and physiology. For

example, large-seeded B. grandis germinated many seedlings, but few survived to the end of the

experiment. S. tarapotanum and Casearia sp. nov., on the other hand, recruited at unexpectedly

high densities, despite germinating from relatively small seeds.

This is the first experimental evidence that mammalian predation contributes positively

to the maintenance of tree species diversity. By preferentially consuming seeds of common and

large-seeded species, rodents increased diversity and generated some of the negative frequency

dependence observed in the recruitment of tropical trees (Harms et al. 2000). To evaluate whether

abundance-biased predation also occurs in the seedling stage as well as in the seed stage, it would

85

be necessary to conduct experimental transplantations of seedlings into exclosures (cf. DeMattia

et al. 2006). Beckage and Clark (2005), in the only previous study to relate mammalian predation

to seedling diversity, found that predation uniformly reduced the recruitment of temperate trees

that varied in competitive ability. Because they studied only three species, they were unable to

test predation’s effect on species richness or relative abundance.

Contributions to Recruitment Limitation

Does the variance in seedling recruitment generated by predation arise primarily in

the seed stage or in the seedling stage? In this study, mammals had stronger and more variable

effects on seed survival than on seedling survival. The pattern of recruited seedling density thus

mirrored that of seed survival with only minor modifications occurring during seedling survival.

Although our data on seedling growth were unbalanced (due to variation in seed survival), growth

rates were not affected by exclosure treatment, adult abundance or seed mass. This may be

explained by the scarcity of mammalian herbivores, relative to that of seed predators, at EBCC.

The present data suggest that patterns of seedling recruitment are primarily driven by differences

in seed survival, which persist through the seedling stage.

CONCLUSIONS

Mammalian predation on seeds and seedlings strongly affects the recruitment of tree

seedlings in tropical forests. Small and medium mammals significantly reduce the density of

recruiting seedlings, while increasing recruit diversity. They increase per-stem species richness

by increasing the relative abundance of rare and small-seeded species. Large mammals have

minimal effects on the recruitment of tree seedlings. Predation – especially by rodents – plays an

important role in maintaining tree diversity in tropical forests.

86

CHAPTER 5 – CONCLUSIONS

SUMMARY

The objective of this dissertation has been to contribute to our understanding of the

factors that affect tree community structure in tropical forests through a series of manipulative

experiments on seeds and seedlings. Though there are many factors that affect community struc-

ture, the foci of this dissertation are three of the most important: seed dispersal, competition and

predation. This study breaks new ground in three ways. No experimental study had partitioned

the variance in seedling community structure in the tropics. Chapter 2 showed seed dispersal to

be more important than environmental conditions in explaining not only stem density, but also

species richness and composition. This suggests that theories of neutrality may be sufficient to

explain community structure in tropical forests, at least for young plants, and at least at small spa-

tial scales. Few previous studies had assessed the intensity of competition among seedlings as we

did in Chapter 3. Our results essentially ruled out inter-seedling competition as a factor structur-

ing the seedling layer. Predation by mammals has long been seen as a significant limit to seedling

recruitment. Chapter 4 built on previous results with a three-year predator-exclusion experiment

including many focal species. We showed that small mammals are the most important seed preda-

tors, and that they can increase seedling diversity by generating a rare-species advantage (Paine

and Beck in press).

SYNTHESIS

How do seed dispersal and environmental variation shape seedling recruitment in tropical

forests? The environmental variation that I contrasted against dispersal treatments included com-

petition and predation, among other processes. The results of Chapter 2 indicate that the patterns

87

generated by seed dispersal shape community structure more strongly than do environmental con-

ditions. Of the three processes examined in this dissertation, dispersal thus appeared to be more

important in structuring the seedling layer than is competition or predation, at least initially.

Looking to the environmental factors that affect post-dispersal establishment success,

which are most important? Tolerance of intense asymmetric competition with adult trees is criti-

cal to the survival of seedlings in the understory of tropical forests; competition with adults is

closely linked with the availability of light (Lewis and Tanner 2000). The results from Chapter 3

show that inter-seedling competition adds little to this relationship. Seedlings evidently compete

weakly, if at all, among themselves; the competitive milieu of seedlings is dominated by adult

trees.

Predation by terrestrial mammals, on the other hand, imposes a strong ecological filter,

removing many seeds and seedlings from the community. Terrestrial mammals, particularly

rodents, selectively remove and consume large-seeded and common species, a distinctly non-

neutral process. The seeds sown in the experiments of Chapter 2 were all exposed to mammalian

predators, and yet dispersal explained more of the variance in community structure than did

environmental conditions (which included predation). This suggests that while predation is an

important aspect of post-dispersal ecological filtration, it may be less important than dispersal.

How will the relative importance of dispersal, competition and predation change through

time as seedlings grow and age? The relationships between dispersal treatments and seedling

community structure became weaker over the duration of Chapter 2’s seed-addition experiment

(Figures 2.4 and 2.6). On the other hand, the effects of environmental variation on community

structure are likely to intensify through time. This is particularly true of competition. Competition

may be weak among seedlings, but is likely to intensify as their stature increases (Uriarte et al.

88

2004). The effects of predation on community structure, on the other hand, are likely to diminish

as plants grow. We found seed predation to be more important than seedling herbivory (Paine

and Beck in press). The likelihood that herbivores may kill seedlings declines as they increase in

size, as their leaf area and ability to resprout following damage increase (Barone 2000). Even as

the overall importance of environmental variables increase, the relative importance of competi-

tion and predation will shift through time. Further, and longer-term, studies will be necessary to

evaluate how the relative importance of these factors will shift through time.

89

LITERATURE CITED

Asquith, N. M., and M. Mejia-Chang. 2005. Mammals, edge effects, and the loss of tropical forest diversity. Ecology 86:379-390.

Asquith, N. M., S. J. Wright, and M. J. Clauss. 1997. Does mammal community composition control recruitment in Neotropical forests? Evidence from Panama. Ecology 78:941-946.

Augspurger, C. K. 1984. Seedling survival of tropical tree species: interactions of dispersal distance, light gaps, and pathogens. Ecology 65:1705-1712.

Baiker, J. 2007. Variation in the density and diversity of seed rain in two neotropical rainforests. Swiss Federal Institute of Technology, Zurich, Switzerland.

Barbaris, I. M., and E. V. J. Tanner. 2005. Gaps and root trenching increase tree seedling growth in Panamanian semi-evergreen forest. Ecology 86:667-674.

Barone, J. A. 2000. Comparison of herbivores and herbivory in the canopy and understory for two tropical tree species. Biotropica 32:307-317.

Beck, H. 2005. Seed predation and dispersal by peccaries throughout the Neotropics and its consequences: a review and synthesis. Pages 77-115 in P. M. Forget, J. E. Lambert, P. E. Hulme, and S. B. Vander Wall, editors. Seed fate: predation, dispersal and seedling establishment. CAB International, Wallingford, UK.

Beck, H. 2006. A review of peccary - palm interactions and their ecological ramifications across the Neotropics. Journal of Mammology 87:519-530.

Beck, H., M. S. Gaines, J. E. Hines, and J. D. Nichols. 2004. Comparative dynamics of small mammal populations in treefall gaps and surrounding understorey within Amazonian rainforest. Oikos 106:27-38.

Beck, H., and J. Terborgh. 2002. Groves versus isolates: how spatial aggregation of Astrocaryum murumuru palms affects seed removal. Journal of Tropical Ecology 18:275-288.

Beck-King, H., O. v. Helversen, and R. Beck-King. 1999. Home range, population density, and food resources of Agouti paca (Rodentia: Agoutidae) in Costa Rica: A study using alternative methods. Biotropica 31:675-685.

90

Beckage, B., and J. S. Clark. 2005. Does predation contribute to tree diversity? Oecologia 143:458-469.

Bell, G. 2001. Neutral Macroecology. Science 293:2413-2418.

Bell, T., R. P. Freckleton, and O. T. Lewis. 2006. Plant pathogens drive density-dependent seedling mortality in a tropical tree. Ecology Letters 9:569-574.

Bloor, J. M. G., and P. J. Grubb. 2003. Growth and mortality in high and low light: trends among 15 shade-tolerant tropical rain forest tree species. Journal of Ecology 91:77-85.

Blundell, A. G., and D. R. Peart. 1998. Distance-dependence in herbivory and foliar condition for juvenile Shorea trees in Bornean dipterocarp rain forest. Oecologia 117:151-160.

Bodmer, R. E., and D. Ward. 2006. Frugivory in large mammalian herbivores. Pages 232-260 in K. Danell, R. Bergström, P. Duncan, and J. Pastor, editors. Large herbivore ecology, ecosystem dynamics and conservation. Cambridge University Press.

Brewer, S. W. 2001. Predation and dispersal of large and small seeds of a tropical palm. Oikos 92:245-255.

Brown, N. D., and T. C. Whitmore. 1992. Do dipterocarp seedlings really partition rainforest gaps? Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 335:369-378.

Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Inference: A Practical Information-Theoretic Approach. 2nd edition. Springer.

Burslem, D. F. R. P., P. J. Grubb, and I. M. Turner. 1995. Responses to Nutrient Addition among Shade-Tolerant Tree Seedlings of Lowland Tropical Rain-Forest in Singapore. Journal of Ecology 83:113-122.

Canham, C. D., J. S. Denslow, W. J. Platt, J. R. Runkle, T. A. Spies, and P. S. White. 1990. Light regimes beneath closed canopies and tree-fall gaps in temperate and tropical forests. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 20:620-631.

Capers, R. S., R. L. Chazdon, A. Redondo Brenes, and B. Vilchez Alvarado. 2005. Successional dynamics of woody seedling communities in wet tropical secondary forests. Journal of

91

Ecology 93:1071-1084.

Casper, B. B., and R. B. Jackson. 1997. Plant competition underground. Annual Review of Ecology and Systematics 28:545-570.

Casper, B. B., H. J. Schenk, and R. B. Jackson. 2003. Defining a plant’s below ground zone of influence. Ecology 84:2313-2321.

Castellano, S., and E. Balletto. 2002. Is the partial Mantel test inadequate? Evolution 56:1871-1873.

Caswell, H. 1976. Community structure - neutral model analysis. Ecological Monographs 46:327-354.

Chase, J. M. 2003. Community assembly: when should history matter? Oecologia 136:489-498.

Chase, J. M., and M. A. Leibold. 2003. Ecological Niches: linking classical and contemporary approaches. University of Chicago Press, Chicago.

Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366.

Clark, D. B., and D. A. Clark. 1989. The role of physical damage in the seedling mortality regime of a neotropical rain forest. Oikos 55:225-230.

Clark, J. S., E. Macklin, and L. Wood. 1998. Stages and spatial scales of recruitment limitation in southern Appalachian forests. Ecological Monographs 68:213-235.

Clark, J. S., and J. S. McLachlan. 2003. Stability of forest biodiversity. Nature 423:635-638.

Clark, J. S., M. Silman, R. Kern, E. Macklin, and J. HilleRisLambers. 1999. Seed dispersal near and far: Patterns across temperate and tropical forests. Ecology 80:1475-1494.

Coley, P. D. 1983. Herbivory and Defensive Characteristics of Tree Species in a Lowland Tropical Forest. Ecological Monographs 53:209-233.

Coley, P. D., and J. A. Barone. 1996. Herbivory and plant defenses in tropical forests. Annual Review of Ecology and Systematics 27:305-335.

92

Condit, R., P. S. Ashton, P. Baker, S. Bunyavejchewin, S. Gunatilleke, N. Gunatilleke, S. P. Hubbell, R. B. Foster, A. Itoh, J. V. LaFrankie, H. S. Lee, E. Losos, N. Manokaran, R. Sukumar, and T. Yamakura. 2000. Spatial patterns in the distribution of tropical tree species. Science 288:1414-1418.

Condit, R., N. Pitman, E. G. Leigh, J. Chave, J. Terborgh, R. B. Foster, P. Nunez, S. Aguilar, R. Valencia, G. Villa, H. C. Muller-Landau, E. Losos, and S. P. Hubbell. 2002. Beta-diversity in tropical forest trees. Science 295:666-669.

Connell, J. H. 1971. On the role of natural enemies in preventing competitive exclusion in some marine animals and in rain forest trees. Pages 289-312 in P. J. den Boer and G. R. Gradwell, editors. Dynamics of numbers in populations. Center for Agricultural Publication and Documentation, Osterbeek Wageningen, The Netherlands.

Connell, J. H. 1978. Diversity in Tropical Rain Forests and Coral Reefs. Science 199:1302-1310.

Connell, J. H. 1980. Diversity and the coevolution of competitors, or the ghost of competition past. Oikos 35:131-138.

Connell, J. H., I. Debski, C. A. Gehring, L. Goldwasser, P. T. Green, K. E. Harms, P. Juniper, and T. C. Theimer. 2005. Dynamics of seedling recruitment in an Australian tropical rainforest. Pages 486-506 in E. Bermingham, C. W. Dick, and C. Moritz, editors. Tropical rainforests: Past, present and future. University of Chicago Press, Chicago.

Connell, J. H., T. P. Hughes, and C. C. Wallace. 1997. A 30-year study of coral abundance, recruitment, and disturbance at several scales in space and time. Ecological Monographs 67:461-488.

Coomes, D. A., and P. J. Grubb. 2000. Impacts of root competition in forests and woodlands: A theoretical framework and review of experiments. Ecological Monographs 70:171-207.

Croat, T. B. 1978. Flora of Barro Colorado Island. Stanford University Press, Stanford, California.

Curran, L. M., and M. Leighton. 2000. Vertebrate responses to spatiotemporal variation in seed production of mast-fruiting Dipterocarpaceae. Ecological Monographs 70.

Curran, L. M., and C. O. Webb. 2000. Experimental tests of the spatiotemporal scale of seed predation in mast-fruiting Dipterocarpaceae. Ecological Monographs 70:129-148.

93

Dalling, J. W., and K. E. Harms. 1999. Damage tolerance and cotyledonary resource use in the tropical tree Gustavia superba. Oikos 85:257-264.

Dalling, J. W., and S. P. Hubbell. 2002. Seed size, growth rate and gap microsite conditions as determinants of recruitment success for pioneer species. Journal of Ecology 90:557-568.

Dalling, J. W., H. C. Muller-Landau, S. J. Wright, and S. P. Hubbell. 2002. Role of dispersal in the recruitment limitation of neotropical pioneer species. Journal of Ecology 90:714-727.

Davies, S. D., S. J. M. Droop, P. Gregerson, L. Henson, C. J. Leon, J. L. Villa-Lobos, H. Synge, and J. Zantovska. 1986. Plants in Danger: what do we know. International Union of the Conservation of Nature, Gland, Switzerland.

DeMattia, E. A., L. M. Curran, and B. J. Rathcke. 2004. Effects of small rodents and large mammals on Neotropical seeds. Ecology 85:2161-2170.

DeMattia, E. A., B. J. Rathcke, L. M. Curran, A. R., and O. Vargas. 2006. Effects of small rodent and large mammal exclusion on seedling recruitment in Costa Rica. Biotropica 38:196-202.

Denslow, J. S. 1980. Gap Partitioning among Tropical Rainforest Trees. Biotropica 12:47-55.

Denslow, J. S., E. Newell, and A. M. Ellison. 1991. The Effect of Understory Palms and Cyclanths on the Growth and Survival of Inga Seedlings. Biotropica 23:225-234.

Denslow, J. S., J. C. Schultz, P. M. Vitousek, and B. R. Strain. 1990. Growth-Responses of Tropical Shrubs to Treefall Gap Environments. Ecology 71:165-179.

Emmons, L. H. 1982. Ecology of Proechimys (Rodentia, Echimyidae) in Southeastern Peru. Tropical Ecology 23:280-290.

Engelbrecht, B. M. J., and T. A. Kursar. 2003. Comparative drought-resistance of seedlings of 28 species of co-occurring tropical woody plants. Oecologia 136:383-393.

Engelbrecht, B. M. J., T. A. Kursar, and M. T. Tyree. 2005. Drought effects on seedling survival in a tropical moist forest. Trees-Structure and Function 19:312-321.

94

Fargione, J., C. S. Brown, and D. Tilman. 2003. Community assembly and invasion: An experimental test of neutral versus niche processes. Proceedings of the National Academy of Sciences of the United States of America 100:8916-8920.

Forget, P. M., K. Kitajima, and R. B. Foster. 1999. Pre- and post-dispersal seed predation in Tachigali versicolor (Caesalpiniaceae): effects of timing of fruiting and variation among trees. Journal of Tropical Ecology 15:61-81.

Foster, B. L., T. L. Dickson, C. A. Murphy, I. S. Karel, and V. H. Smith. 2004. Propagule pools mediate community assembly and diversity-ecosystem regulation along a grassland productivity gradient. Journal of Ecology 92.

Foster, B. L., and D. Tilman. 2003. Seed limitation and the regulation of community structure in oak savanna grassland. Journal of Ecology 91:999-1007.

Foster, R. B. 1990. The Floristic Composition of the Rio Manu Floodplain. Pages 99-111 in A. H. Gentry, editor. Four Neotropical Rainforests. Yale University Press, New Haven.

Foster, S. A., and C. H. Janson. 1985. The relationship between seed size and establishment conditions in tropical woody plants. Ecology 66:773-780.

Frazer, G. W., C. D. Canham, and K. P. Lertzman. 2000. Gap light analyzer (GLA), Version 2.0: image-processing software to analyze true-color, hemispherical canopy photographs. Bulletin of the Ecological Society of America 81.

Fukami, T. 2004a. Assembly history interacts with ecosystem size to influence species diversity. Ecology 85:3234-3242.

Fukami, T. 2004b. Community assembly along a species pool gradient: implications for multiple-scale patterns of species diversity. Population Ecology 46:137-147.

Garwood, N. C. 1983. Seed-Germination in a Seasonal Tropical Forest in Panama - a Community Study. Ecological Monographs 53:159-181.

Gentry, A. H. 1988. Changes in Plant Community Diversity and Floristic Compositio on Environmental and Geographical Gradients. Annals of the Missouri Botanical Garden 75:1-34.

Gentry, A. H. 1990. Four Neotropical Forests. Yale University Press, New Haven, CT.

95

Gilbert, B., and M. J. Lechowicz. 2004. Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences 101:7651-7656.

Grubb, P. J. 1977. Control of Forest Growth and Distribution on Wet Tropical Mountains: With Special Reference to Mineral Nutrition. Annual Review of Ecology and Systematics 8:83-107.

Hammond, D. S., and V. K. Brown. 1998. Disturbance, phenology and life-history characteristics: factors influencing distance/density-dependent attack on tropical seeds and seedlings. Pages 51-78 in D. M. Newbery, H. H. T. Prins, and N. D. Brown, editors. Dynamics of Tropical Communities. Blackwell, Oxford.

Harms, K. E., and J. W. Dalling. 1997. Damage and herbivory tolerance through resprouting as an advantage of large seed size in tropical trees and lianas. Journal of Tropical Ecology 13:617-621.

Harms, K. E., and C. E. T. Paine. 2003. Regeneración de árboles tropicales e implicaciones para el manejo de bosques naturales. Ecosistemas 12.

Harms, K. E., J. S. Powers, and R. A. Montgomery. 2004. Variation in small sapling density, understory cover and resource availability in four Neotropical forests. Biotropica 36:40-51.

Harms, K. E., S. J. Wright, O. Calderon, A. Hernandez, and E. A. Herre. 2000. Pervasive density-dependent recruitment enhances seedling diversity in a tropical forest. Nature 404:493-495.

Harper, J. L. 1977. Population biology of plants. Academic Press, London.

Harpole, W. S., and D. Tilman. 2006. Non-neutral patterns of species abundance in grassland communities. Ecology Letters 9:15-23.

Hartshorn, G. S. 1980. Neotropical Forest Dynamics. Biotropica 12:23-30.

He, F. L., P. Legendre, and J. V. LaFrankie. 1997. Distribution patterns of tree species in a Malaysian tropical rain forest. Journal of Vegetation Science 8:105-114.

Hille Ris Lambers, J., J. S. Clark, and B. Beckage. 2002. Density-dependent mortality and the latitudinal gradient in species diversity. Nature 417:732-735.

96

Hoffmann, W. A., and H. Poorter. 2002. Avoiding bias in calculations of relative growth rate. Annals of Botany 90:37-42.

Holdridge, L. R. 1947. Determination of world plant formations from simple climatic data. Science 105:367-368.

Holl, K. D., and M. E. Lulow. 1997. Effects of species, habitat, and distance from edge on post-dispersal seed predation in a tropical rainforest. Biotropica 29:459-468.

Holt, R. D. 1977. Predation, apparent competition, and structure of prey communities. Theoretical Population Biology 12:197-229.

Horn, H. S., R. Nathan, and S. R. Kaplan. 2001. Long-distance dispersal of tree seeds by wind. Ecological Research 16:877-885.

Howlett, B. E., and D. W. Davidson. 2001. Herbivory on planted dipterocarp seedlings in secondary logged forests and primary forests of Sabah, Malaysia. Journal of Tropical Ecology 17:285-302.

Hubbell, S. P. 2001. The unified theory of biodiversity and biogeography. Princeton University Press, Princeton, New Jersey.

Hubbell, S. P. 2006. Neutral theory and the evolution of ecological equivalence. Ecology 87:1387-1398.

Hubbell, S. P., R. B. Foster, S. T. O’Brien, K. E. Harms, R. Condit, B. Wechsler, S. J. Wright, and S. Loo de Lao. 1999. Light gap disturbances, recruitment limitation, and tree diversity in a Neotropical forest. Science 283:554-557.

Hulme, P. E. 2001. Seed-eaters: Seed dispersal, destruction and demography.in D. J. Levey, W. R. Silva, and M. Galetti, editors. Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CABI Publishing, Wallingford, UK.

Huntly, N. 1991. Herbivores and the Dynamics of Communities and Ecosystems. Annual Review of Ecology and Systematics 22:477-503.

Huston, M. A., and D. L. DeAngelis. 1994. Competition and coexistence - the effects of resource transport and supply rates. American Naturalist 144:954-977.

97

Hyatt, L. A., M. S. Rosenberg, T. G. Howard, G. Bole, W. Fang, J. Anastasia, K. Brown, R. Grella, K. Hinman, J. P. Kurdziel, and J. Gurevitch. 2003. The distance dependence prediction of the Janzen-Connell hypothesis: a meta-analysis. Oikos 103:590-602.

Janson, C. H., and L. H. Emmons. 1990. Ecological structure of the nonflying mammal community at Cocha Cashu Biological Station, Manu National Park, Peru. Pages 314-338 in A. H. Gentry, editor. Four Neotropical Forests. Yale University Press, New Haven.

Janzen, D. H. 1970. Herbivores and Number of Tree Species in Tropical Forests. American Naturalist 104:501.

John, R., J. W. Dalling, K. E. Harms, J. B. Yavitt, R. F. Stallard, M. Mirabello, S. P. Hubbell, R. Valencia, H. Navarrete, M. Vallejo, and R. B. Foster. 2007. Soil nutrients influence spatial distributions of tropical tree species. Proceedings of the National Academy of Sciences 104:864-869.

Karst, J., B. Gilbert, and M. J. Lechowicz. 2005. Fern community assembly: the roles of chance and the environment at local and intermediate scales. Ecology 86:2473-2486.

Kimura, M. 1983. The neutral theory of molecular evolution. Cambridge University Press, Cambridge.

King, D. A. 1994. Influence of Light Level on the Growth and Morphology of Saplings in a Panamanian Forest. American Journal of Botany 81:948-957.

Kitajima, K. 1994. Relative Importance of Photosynthetic Traits and Allocation Patterns as Correlates of Seedling Shade Tolerance of 13 Tropical Trees. Oecologia 98:419-428.

Klein, J. P., and M. L. Moeschberger. 1997. Survival analysis techniques for censored and truncated data. Springer-Verlag, New York.

Klironomos, J. N. 2002. Feedback with soil biota contributes to plant rarity and invasiveness in communities. Nature 417:67-70.

Kobe, R. K. 1999. Light gradient partitioning among tropical tree species through differential seedling mortality and growth. Ecology 80:187-201.

98

Latham, R. E., and R. E. Ricklefs. 1993. Continental comparisons or temperate-zone tree species diversity. Pages 294-314 in R. E. Ricklefs and D. Schluter, editors. Species Diversity in Ecological Communities. University of Chicago Press, Chicago.

Lawrence, D. 2003. The response of tropical tree seedlings to nutrient supply: meta-analysis for understanding a changing tropical landscape. Journal of Tropical Ecology 19:239-250.

Leibold, M. A., and M. A. McPeek. 2006. Coexistence of the niche and neutral perspectives in community ecology. Ecology 87:1399-1410.

Levin, S. A., H. C. Muller-Landau, R. Nathan, and J. Chave. 2003. The Ecology and Evolution of Seed Dispersal: A Theoretical Perspective. Annual Review of Ecology Evolution and Systematics 34:575-604.

Levine, J. M., and D. J. Murrell. 2003. The community-level consequences of seed dispersal patterns. Annual Review of Ecology Evolution and Systematics 34:549-574.

Lewis, S. L., and E. V. J. Tanner. 2000. Effects of above- and below ground competition on growth and survival of rain forest tree seedlings. Ecology 81:2525-2538.

Lieberman, D., M. Lieberman, R. Peralta, and G. S. Hartshorn. 1985. Mortality Patterns and Stand Turnover Rates in a Wet Tropical Forest in Costa-Rica. Journal of Ecology 73:915-924.

Littell, R. C., W. W. Stroup, and R. J. Freund. 2002. SAS for Linear Models. SAS Publishing.

Makana, J.-R., and S. C. Thomas. 2004. Dispersal limits natural recruitment of African mahoganies. Oikos 106:67-72.

Mantel, N. 1967. The detection of disease clustering and a generalized regression approach. Cancer Res 27:209-220.

Marks, P. L., and P. A. Harcombe. 1981. Forest Vegetation of the Big Thicket, Southeast Texas. Ecological Monographs 51:287-305.

Marquis, R. J., H. J. Young, and H. E. Braker. 1986. The influence of understory vegetation cover on germination and seedling establishment in a tropical lowland wet forest. Biotropica 18:273-278.

99

Massey, F. P., K. Massey, M. C. Press, and S. E. Hartley. 2006. Neighbourhood composition determines growth, architecture and herbivory in tropical rain forest tree seedlings. Journal of Ecology 94:646.

McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software Design.

McGill, B. J. 2003. A test of the unified neutral theory of biodiversity. Nature 422:881-885.

McGill, B. J., B. A. Maurer, and M. D. Weiser. 2006. Empirical evaluation of neutral theory. Ecology 87:1411-1423.

Moles, A. T., D. I. Warton, and M. Westoby. 2003. Do small-seeded species have higher survival through seed predation than large-seeded species? Ecology 84:3148-3161.

Moles, A. T., and M. Westoby. 2004a. Seedling survival and seed size: a synthesis of the literature. Journal of Ecology 92:372-383.

Moles, A. T., and M. Westoby. 2004b. What do seedlings die from and what are the implications for evolution of seed size? Oikos 106:193-199.

Molofsky, J., and C. K. Augspurger. 1992. The effect of leaf litter on early seedling establishment in a tropical forest. Ecology 73:68-77.

Montgomery, R. A., and R. L. Chazdon. 2002. Light gradient partitioning by tropical tree seedlings in the absence of canopy gaps. Oecologia 131:165-174.

Moora, M., R. Sen, and M. Zobel. 2004. Native arbuscular mycorrhizal fungal communities differentially influence the seedling performance of rare and common Pulsatilla species. Functional Ecology 18:554-562.

Moore, K. A., and S. C. Elmendorf. 2006. Propagule vs. niche limitation: untangling the mechanisms behind plant species distributions. Ecology Letters 9:797-804.

Muller-Landau, H. C., J. W. Dalling, K. E. Harms, S. J. Wright, R. Condit, S. P. Hubbell, and R. B. Foster. 2004. Seed dispersal and density-dependent seed and seedling mortality in Trichilia tuberculata and Miconia argentea. Pages 340-362 in E. C. Losos and E. G. Leigh, editors. Forest Diversity and Dynamism: Findings from a Network of Large-Scale Tropical Forest Plots. University of Chicago Press, Chicago.

100

Muller-Landau, H. C., R. B. Foster, S. J. Wright, O. Calderon, and S. Hubbell. 2002. Assessing Recruitment Limitation: Concepts, Methods, and Case-Studies from a Tropical Forest.in D. J. Levey, W. R. Silva, and M. Galletti, editors. Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CABI International, Wallingford, UK.

Munzbergova, Z., and T. Herben. 2005. Seed, dispersal, microsite, habitat and recruitment limitation: identification of terms and concepts in studies of limitations. Oecologia 145:1-8.

Nathan, R., G. G. Katul, H. S. Horn, S. M. Thomas, R. Oren, R. Avissar, S. W. Pacala, and S. A. Levin. 2002. Mechanisms of long-distance dispersal of seeds by wind. Nature 418:409-413.

Nathan, R., and H. C. Muller-Landau. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends in Ecology & Evolution 15:278-285.

Nicotra, A. B., R. L. Chazdon, and S. V. B. Iriarte. 1999. Spatial heterogeneity of light and woody seedling regeneration in tropical wet forests. Ecology 80:1908-1926.

Norden, N., J. Chave, A. Caubère, P. Châtelet, N. Ferroni, P.-M. Forget, and C. Thébaud. In Press. Is seedling dynamics determined by environment or by seed arrival? A test in a neotropical forest. Journal of Ecology.

Norghauer, J., J. Malcolm, B. Zimmerman, and J. Felfili. 2006. An experimental test of density- and distant-dependent recruitment of mahogany (Swietenia macrophylla) in southeastern Amazonia. Oecologia 148:437-446.

Notman, E., and D. L. Gorchov. 2001. Variation in post-dispersal seed predation in mature Peruvian lowland tropical forest and fallow agricultural sites. Biotropica 33:621-636.

Notman, E., D. L. Gorchov, and F. H. Cornejo. 1996. Effect of distance, aggregation and habitat on levels of seed predation for two mammal-dispersed Neotropical rain forest tree species. Oecologia 106:221-227.

Novotny, V., Y. Basset, S. E. Miller, G. D. Weiblen, B. Bremer, L. Cizek, and P. Drozd. 2002. Low host specificity of herbivorous insects in a tropical forest. Nature 416:841-844.

Obeid, M., D. Machin, and J. L. Harper. 1967. Influence of density on plant to plant variation in fiber flax, Linum usitatissimum L. Crop Science 7:471-473.

101

Oksanen, J., R. Kindt, and R. B. O’Hara. 2005. vegan: Community Ecology Package.

Olmsted, I., H. Dunevitz, and W. J. Platt. 1993. Effects of freezes on tropical trees in Everglades National Park Florida, USA. Tropical Ecology 34:17-34.

Osher, L. J., and S. W. Buol. 1998. Relationship of soil properties to parent material and landscape position in eastern Madre de Dios, Peru. Geoderma 83:143-166.

Ostertag, R. 1998. Below ground effects of canopy gaps in a tropical wet forest. Ecology 79:1294-1304.

Ostfeld, R. S., R. H. Manson, and C. D. Canham. 1997. Effects of Rodents on Survival of Tree Seeds and Seedlings Invading Old Fields. Ecology 78:1531-1542.

Osunkoya, O. O., J. E. Ash, A. W. Graham, and M. S. Hopkins. 1993. Growth of tree seedlings in tropical rain forests of North Queensland, Australia. Journal of Tropical Ecology 9:1-18.

Pacala, S. W. 1988. Competitive equivalence - the coevolutionary consequences of sedentary habit. American Naturalist 132:576-593.

Pacala, S. W., C. D. Canham, J. Saponara, J. A. Silander, R. K. Kobe, and E. Ribbens. 1996. Forest models defined by field measurements: Estimation, error analysis and dynamics. Ecological Monographs 66:1-43.

Paine, C. E. T., and H. Beck. in press. Seed predation by Neotropical rainforest mammals increases diversity in seedling recruitment. Ecology.

Paz, H., and M. Martinez-Ramos. 2003. Seed mass and seedling performance within eight species of Psychotria (Rubiaceae). Ecology 84:439-450.

Peres, C. A., and C. Baider. 1997. Seed dispersal, spatial distribution and population structure of Brazilnut trees (Bertholletia excelsa) in southeastern Amazonia. Journal of Tropical Ecology 13:595-616.

Peters, H. A. 2003. Neighbour-regulated mortality: the influence of positive and negative density dependence on tree populations in species-rich tropical forests. Ecology Letters 7:757-765.

Pitman, N. C. A., J. Terborgh, M. R. Silman, and P. Nuñez. 1999. Tree species distributions in an upper Amazonian forest. Ecology 80:2651-2661.

102

Pitman, N. C. A., J. W. Terborgh, M. R. Silman, P. Nunez, D. A. Neill, C. E. Ceron, W. A. Palacios, and M. Aulestia. 2001. Dominance and distribution of tree species in upper Amazonian terra firme forests. Ecology 82:2101-2117.

Platt, W. J., and J. H. Connell. 2003. Natural disturbances and directional replacement of species. Ecological Monographs 73:507-522.

Plotkin, J. B., J. M. Chave, and P. S. Ashton. 2002. Cluster analysis of spatial patterns in Malaysian tree species. American Naturalist 160:629-644.

Potts, M. D., P. S. Ashton, L. S. Kaufman, and J. B. Plotkin. 2002. Habitat patterns in tropical rain forests: A comparison of 105 plots in Northwest Borneo. Ecology 83:2782-2797.

R Development Core Team. 2006. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Rasband, W. S. 1997-2005. ImageJ. U. S. National Institutes of Health, Bethesda, Maryland, USA.

Roldan, A. I., and J. A. Simonetti. 2001. Plant-mammal interactions in tropical Bolivian forests with different hunting pressures. Conservation Biology 15:617-623.

Rosenzweig, M. L. 1995. Species Diversity in Space and Time. Cambridge University Press.

Russo, S. E. 2003. Responses of dispersal agents to tree and fruit traits in Virola calophylla (Myristicaceae): implications for selection. Oecologia 136:80-87.

Russo, S. E. 2005. Linking seed fate to natural dispersal patterns: factors affecting predation and scatter-hoarding of Virola calophylla seeds in Peru. Journal of Tropical Ecology 21:243-253.

Russo, S. E., and C. K. Augspurger. 2004. Aggregated seed dispersal by spider monkeys limits recruitment to clumped patterns in Virola calophylla. Ecology Letters 7:1058-1067.

Sala, O. E., R. A. Golluscio, W. K. Lauenroth, and A. Soriano. 1989. Resource partitioning between shrubs and grasses in the Patagonian steppe. Oecologia 81:510-505.

Saverimuttu, T., and M. Westoby. 1996. Seedling longevity under deep shade in relation to seed size. Journal of Ecology 84:681-689.

103

Schnitzer, S. A., and W. P. Carson. 2001. Treefall gaps and the maintenance of species diversity in a tropical forest. Ecology 82:913-919.

Schnitzer, S. A., M. Kuzee, and F. Bongers. 2005. Disentangling above-and below-ground competition between lianas and trees in a tropical forest. Journal of Ecology 93:1115:1125.

Schoener, T. W. 1971. Theory of feeding strategies. Annual Review of Ecology and Systematics 2:369-404.

Schupp, E. W., T. Milleron, and S. E. Russo. 2002. Dissemination Limitation and the Origin and Maintenance of Species-rich Tropical Forests.in D. J. Levey, W. R. Silva, and M. Galetti, editors. Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CAB International.

Seidler, T. G., and J. B. Plotkin. 2006. Seed dispersal and spatial pattern in tropical trees. Plos Biology 4:2132-2137.

Silman, M. 1996. Regeneration from seed in a neotropical rain forest. Duke Univ., Durham, NC.

Silman, M. R., J. W. Terborgh, and R. A. Kiltie. 2003. Population regulation of a dominant-rain forest tree by a major seed-predator. Ecology 84:431-438.

Smouse, P. E., J. C. Long, and R. R. Sokal. 1986. Multiple Regression and Correlation Extensions of the Mantel Test of Matrix Correspondence. Systematic Zoology 35:627-632.

Stohlgren, T. 1993. Intra-specific competition (crowding) of giant sequoias (Sequoiadendron giganteum). Forest Ecology and Management 59:127-148.

Svenning, J. C., and S. J. Wright. 2005. Seed limitation in a Panamanian forest. Journal of Ecology 93:853.

Tanner, E. V. J., V. K. Teo, D. A. Coomes, and J. J. Midgley. 2005. Pair-wise competition-trials amongst seedlings of ten dipterocarp species; the role of initial height, growth rate and leaf attributes. Journal of Tropical Ecology 21:317-328.

Taylor, K. M., and L. W. Aarssen. 1989. Neighbor effects in mast year seedlings of Acer saccharum. American Journal of Botany 76:546-554.

104

Terborgh, J. 1983. Five New World primates: a study in comparative ecology. Princeton University Press, Princeton, New Jersey, USA.

Terborgh, J. 1990. An overview of research at Cocha Cashu Biological Station. Pages 48-59 in A. H. Gentry, editor. Four Neotropical Forests. Yale University Press, New Haven, CT.

Terborgh, J. 1999. Requiem for Nature. Island Press, Washington, D. C.

Terborgh, J., L. Lopez, P. Nunez, M. Rao, G. Shahabuddin, G. Orihuela, M. Riveros, R. Ascanio, G. H. Adler, T. D. Lambert, and L. Balbas. 2001. Ecological meltdown in predator-free forest fragments. Science 294:1923-1926.

Terborgh, J., E. Losos, M. P. Riley, and M. B. Riley. 1993. Predation by vertebrates and invertebrates on the seeds of 5 canopy tree species of an Amazonian forest. Vegetatio 108:375-386.

Terborgh, J., G. Nuñez-Iturri, A. J. Pitman, F. Cornejo, P. Alvarez, V. Swamy, B. Pringle, and C. E. T. Paine. in review. Tree Recruitment in an “empty” forest. Ecology.

Terborgh, J., S. K. Robinson, T. A. P. Iii, C. A. Munn, and N. Pierpont. 1990. Structure and Organization of an Amazonian Forest Bird Community. Ecological Monographs 60:213-238.

Tilman, D. 1982. Resource Competition and Community Structure. Princeton University Press, Princeton, New Jersey.

Tilman, D. 1994. Competition and biodiversity in spatially structured habitats. Ecology 75:2-16.

Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78:81-92.

Tuomisto, H., K. Ruokolainen, and M. Yli-Halla. 2003. Dispersal, environment, and floristic variation of western Amazonian forests. Science 299:241-244.

Turnbull, L. A., M. J. Crawley, and M. Rees. 2000. Are plant populations seed-limited? A review of seed sowing experiments. Oikos 88:225-238.

105

Uriarte, M., C. D. Canham, J. Thompson, J. K. Zimmerman, and N. Brokaw. 2005. Seedling recruitment in a hurricane-driven tropical forest: light limitation, density-dependence and the spatial distribution of parent trees. Journal of Ecology 93:291-304.

Uriarte, M., R. Condit, C. D. Canham, and S. P. Hubbell. 2004. A spatially explicit model of sapling growth in a tropical forest: does the identity of neighbours matter? Journal of Ecology 92:348-360.

Vander Wall, S. B., K. M. Kuhn, and M. J. Beck. 2005. Seed removal, seed predation and secondary dispersal. Ecology 86:801-806.

Veenendaal, E. M., M. D. Swaine, R. T. Lecha, M. F. Walsh, I. K. Abebrese, and K. Owusu-Afriyie. 1996. Responses of West African forest tree seedlings to irradiance and soil fertility. Functional Ecology 10:501-511.

Venables, W. N., and B. D. Ripley. 1999. Modern applied statistics with S-Plus. Springer New York.

Vieira, E. M., M. A. Pizo, and P. Izar. 2003. Fruit and seed exploitation by small rodents of the Brazilian Atlantic forest. Mammalia 67:533-540.

Volkov, I., J. R. Banavar, F. L. He, S. P. Hubbell, and A. Maritan. 2005. Density dependence explains tree species abundance and diversity in tropical forests. Nature 438:658-661.

Volkov, I., J. R. Banavar, S. P. Hubbell, and A. Maritan. 2003. Neutral theory and relative species abundance in ecology. Nature 424:1035-1037.

Wahungu, G. M., C. P. Catterall, and M. F. Olsen. 2002. Seedling predation and growth at a rainforest-pasture ecotone, and the value of shoots as seedling analogues. Forest Ecology and Management 162:251-260.

Webb, C. O., and D. R. Peart. 1999. Seedling density dependence promotes coexistence of Bornean rain forest trees. Ecology 80:2006-2017.

Webb, C. O., and D. R. Peart. 2001. High seed dispersal rates in faunally intact tropical rain forest: theoretical and conservation implications. Ecology Letters 4:491-499.

106

Wenny, D. G., and D. J. Levey. 1998. Directed seed dispersal by bellbirds in a tropical cloud forest. Proceedings of the National Academy of Sciences of the United States of America 95:6204-6207.

Wills, C., K. E. Harms, R. Condit, D. King, J. Thompson, F. L. He, H. C. Muller-Landau, P. M. S. Ashton, E. Losos, L. Comita, S. Hubbell, J. LaFrankie, S. Bunyavejchewin, H. S. Dattaraja, S. J. Davies, S. Esufali, R. Foster, C. V. S. Gunatilleke, S. Gunatilleke, P. Hall, A. Itoh, R. John, S. Kiratiprayoon, S. Loo de Lao, M. Massa, C. Nath, N. Noor, A. R. Kassim, R. Sukumar, H. S. Suresh, I. F. Sun, S. Tan, T. Yakamura, and J. K. Zimmerman. 2006. Nonrandom Processes Maintain Diversity in Tropical Forests. Science 311:527-531.

Wright, S. J. 2002. Plant diversity in tropical forests: a review of mechanisms of species coexistence. Oecologia 130:1-14.

Yamada, T., E. Suzuki, T. Yamakura, and S. Tan. 2005. Tap-root depth of tropical seedlings in relation to species-specific edaphic preferences. Journal of Tropical Ecology 21:155-160.

Yoda, K., T. Kira, H. Ogawa, and K. Hozumi. 1963. Self-thinning in overcrowded pure stands under cultivated and natural conditions. Journal of Biology 14:107-129.

Zobel, M., M. Otsus, J. Liira, M. Moora, and T. Mols. 2000. Is small-scale species richness limited by seed availability or microsite availability? Ecology 81:3274-3282.

107

APPENDIX 1 – DENSITIES OF SEEDLINGS IN SEED-ADDITION EXPERIMENTS OF CHAPTER 2

All species observed in 572 1-m2 seedling plots during four annual censuses are listed.

Scientific Name Species Code C

Number of Stems Percent of plots‘03 ‘04 ‘05 ‘06 ‘03 ‘04 ‘05 ‘06

AnacardiaceaeSpondias mombin L. SPONMO 3 2 4 1 0 0.8 0.5 0.1 0Tapirira guianensis Aubl. TAPIGU 3 0 1 0 0 0 0.1 0 0

AnnonaceaeAnnonaceae sp. ANNONAC 2 2 4 7 9 0.4 0.6 1 1.2Cremastosperma gracilipes R.E. Fr. CREMGR 3 0 1 1 2 0 0.1 0.1 0.3Cremastosperma leiophyllum (Diels) R.E. Fr. CREMLE 3 3 10 9 6 1.3 1.5 1.5 1Duguetia quitarensis Benth.* DUGUQU 3 3 15 22 17 1.3 2.4 3.2 2.7Guatteria sp. GUATTERIA 2 2 10 3 1 0.4 1.5 0.5 0.1Klarobelia candida Chatrou KLARCA 3 7 18 14 12 2.6 2.7 2.2 2Malmea dielsiana Saff. ex R.E. Fr. MALMDI 3 4 9 14 15 1.7 1.2 2 2.4Oxandra acuminata Diels OXANAC 3 18 68 53 51 7.8 9.7 8.5 8.5Oxandra espintana (Spruce ex Benth.) Baill. OXANES 3 5 9 7 4 1.7 1.3 1.2 0.6Oxandra polyantha R.E. Fr. OXANPO 3 0 1 1 1 0 0.1 0.1 0.1Oxandra sp. OXANDRA 2 1 18 16 8 0.4 3.1 2.7 1.3Porcelia nitida Ruiz & Pav. PORCNI 3 0 0 0 1 0 0 0 0.1Pseudomalmea diclina Chatrou PSEUDI 3 1 5 5 6 0.4 0.8 0.8 1Ruizodendron ovale (Ruiz & Pav.) R.E. Fr. RUIZOV 3 0 7 6 5 0 1.2 0.8 0.8Trigynaea duckei (R.E. Fr.) R.E. Fr. TRIGDU 3 2 4 5 6 0.8 0.6 0.8 1Xylopia cuspidata Diels XYLOCU 3 0 1 1 1 0 0.1 0.1 0.1Xylopia ligustrifolia Humb. & Bonpl. ex

Dunal XYLOPIA 3 7 0 0 0 3 0 0 0

ApocynaceaeApocynaceae sp. APOCYNACEAE 1 2 3 3 4 0.8 0.3 0.3 0.5Aspidosperma megaphyllum Woodson ASPIME 3 2 2 2 1 0.8 0.3 0.3 0.1Forsteronia myriantha Donn. Sm. FOSTMY 3 0 1 2 2 0 0.1 0.3 0.3Landolphia boliviensis Markgr. LANDBO 3 3 9 10 9 1.3 1.2 1.5 1.3Tabernaemontana sananho Ruiz & Pav. TABESA 3 7 16 17 20 3 1.9 2 2.6Tabernaemontana sp. TABERNAEMONTANA 2 1 3 1 1 0.4 0.5 0.1 0.1Tabernaemontana tetrastachya Kunth TABETE 3 0 1 1 1 0 0.1 0.1 0.1

ArecaceaeAstrocaryum murumuru Martius ASTRMU 3 40 80 67 71 15.2 11.8 10.4 10.4Attalea cephalotes Poepp. ex Mart. ATTACE 3 4 14 15 24 1.7 2.4 2.6 4.1Bactris sp. BACTRIS 2 0 1 2 3 0 0.1 0.3 0.5

108



Desmoncus leptospadix Martius DESMLE 3 2 2 3 3 0.4 0.1 0.3 0.3Euterpe precatoria Martius EUTEPR 3 2 7 5 6 0.8 1.2 0.8 1Iriartea deltoidea Ruiz & Pavón* IRIADE 3 20 33 128 93 7.3 4.5 11.9 10.4Oenocarpus bataua Martius OENOBA 3 8 6 6 6 2.1 0.6 0.8 1Socratea exorrhiza (Mart.) H. Wendl.* SOCREX 3 0 0 35 18 0 0 3.4 1.9

BignoniaceaeBignoniaceae ‘funny cots’ BIGNONFC 3 0 2 0 0 0 0.3 0 0Bignoniaceae ‘narrow’ BIGNONNA3 5 14 9 12 2.1 1.9 1.3 1.5Bignoniaceae ‘serrate’ BIGNONSE 3 10 53 41 34 3.9 5.7 5 4.3Bignoniaceae sp. BIGNON 1 19 34 35 30 6.9 5 4.3 4.1Bignoniaceae ‘tapering truncate’ BIGNONTT 3 2 1 1 2 0.8 0.1 0.1 0.3Bignoniaceae ‘truncate 3-veined’ BIGNONT3 3 0 0 1 1 0 0 0.1 0.1Bignoniaceae ‘white stripe’ BIGNONWS 3 0 5 2 2 0 0.6 0.1 0.1Macfadyena unguis-cati (L.) A.H. Gentry MACFUN 3 3 4 2 2 1.3 0.6 0.3 0.3Mansoa sp. MANSOA 2 1 0 0 0 0.4 0 0 0Mansoa standleyi (Steyerm.) A.H. Gentry MANSST 3 8 12 12 10 1.7 1.2 1.2 1.2

BombacaceaeMatisia cordata Bonpl. MATICO 3 0 37 3 22 0 3.1 0.5 1.7Matisia rhombifolia Standl. ex Cuatrec. MATIRH 3 1 8 11 3 0.4 1.3 1.3 0.3Quararibea wittii K. Schum. & Ulbr. QUARWI 3 24 80 71 62 9.5 9.3 8.8 8.5

BoraginaceaeCordia nodosa Lam. CORDNO 3 2 3 2 4 0.8 0.5 0.3 0.6Cordia sp. CORDIA 2 0 1 0 0 0 0.1 0 0

BurseraceaeProtium neglectum Swart PROTNE 3 0 2 1 3 0 0.3 0.1 0.5Protium sp. PROTIUM 2 0 7 3 3 0 0.6 0.5 0.5Protium tenuifolium (Engl.) Engl. PROTTE 3 2 2 2 3 0.8 0.3 0.3 0.5Protium unifoliolatum Engl. PROTUN 3 1 1 1 1 0.4 0.1 0.1 0.1

BuxaceaeStyloceras brokawii A.H. Gentry & R.B.

Foster STYLBR 3 6 13 9 5 2.1 1.2 0.5 0.3

CapparidaceaeCapparis macrophylla Kunth CAPPMA 3 1 1 0 0 0.4 0.1 0 0Capparis sola J.F. Macbr. CAPPSO 3 2 6 6 5 0.8 0.8 0.8 0.8Capparis sp. CAPPARIS 2 0 1 1 2 0 0.1 0.1 0.3Morisonia oblongifolia Britton MORIOB 3 0 1 1 1 0 0.1 0.1 0.1

CecropiaceaePourouma cecropiifolia Martius POURCE 3 2 5 5 3 0.8 0.8 0.8 0.5Pourouma minor Benoist POURMI 3 0 2 1 2 0 0.3 0.1 0.3

109



CelastraceaeMaytenus magnifolia Loes. MAYTMA 3 0 0 0 1 0 0 0 0.1Maytenus sp. MAYTENUS 2 0 4 4 3 0 0.3 0.3 0.1

ChrysobalanaceaeChrysobalanaceae sp. CHRYSOBALANACEAE 1 3 3 3 3 1.3 0.5 0.5 0.5Hirtella lightioides Rusby HIRTLI 3 3 6 5 7 1.3 1 0.8 1.2Licania brittoniana Fritsch LICABR 3 8 24 20 20 2.6 3.6 3.1 3.1Parinari parilis J.F. Macbr. PARIPA 3 0 1 0 0 0 0.1 0 0

ClusiaceaeCalophyllum brasiliense Cambess. CALOBR 3 3 5 2 1 1.3 0.8 0.3 0.1Rheedia acuminata (Ruiz & Pav.) Planch. &

Triana RHEEAC 3 5 8 8 6 2.1 1.2 1.2 0.8

Rheedia brasiliensis Martius RHEEBR 3 3 9 8 7 1.3 1.3 1.2 1Rheedia sp. RHEEDIA 2 0 0 4 7 0 0 0.6 1.2

CombretaceaeBuchenavia grandis Ducke* BUCHGR 3 1 222 55 33 0.4 9.5 3.4 2.7Terminalia oblonga (Ruiz & Pavón) Steudel TERMOB 3 1 5 3 3 0.4 0.5 0.5 0.5Thiloa sp. THILOA 2 0 1 0 0 0 0.1 0 0

ConnaraceaeRourea cuspidata Benth. ex Baker ROURCU 3 2 4 4 5 0.4 0.6 0.6 0.8

ConvulvulaceaeMaripa peruviana Ooststr. MARIPE 3 3 9 11 11 1.3 1.5 1.5 1.5

DichapetalaceaeTapura peruviana K. Krause TAPUPE 3 3 8 9 7 1.3 1.3 1.5 1.2

DilleniaceaeDoliocarpus sp. DOLIOCARPUS 2 2 4 3 3 0.8 0.6 0.5 0.5Tetracera parviflora (Rusby) Sleumer TETRPA 3 0 1 0 0 0 0.1 0 0

EbenaceaeDiospyros pavonii (A. DC.) J.F. Macbr.* DIOSPA 3 1 166 109 60 0.4 8.6 9.3 5.5Diospyros sp. DIOSPYROS 2 1 0 0 0 0.4 0 0 0Diospyros subrotata Hiern* DIOSSU 3 4 12 179 141 1.7 1.9 9.3 9.2

ElaeocarpaceaeSloanea obtusifolia (Moric.) K. Schum. SLOAOB 3 4 33 9 5 1.7 1.3 0.6 0.5Sloanea sp. SLOANEA 2 13 53 29 19 5.2 6.2 4.6 3.1Sloanea terniflora (Sessé & Moc. ex DC.)

Standl. SLOATE 3 1 1 1 1 0.4 0.1 0.1 0.1

EuphorbiaceaeAcalypha diversifolia Jacq. ACALDI 3 0 4 1 2 0 0.1 0.1 0.3Acalypha obovata Benth. ACALOB 3 2 2 2 2 0.4 0.1 0.1 0.1Acalypha sp. ACALYPHA 2 1 6 6 5 0.4 0.8 0.8 0.6

110



Croton tessmannii Mansf. CROTTE 3 4 6 5 7 1.7 1 0.8 1Drypetes amazonica Steyerm. DRYPAM 3 0 0 1 1 0 0 0.1 0.1Drypetes ‘small’ DRYPSM 3 3 6 6 9 1.3 1 1 1.3Mabea maynensis Spruce MABEMA 3 1 55 34 25 0.4 2.2 1.7 1.7Sapium sp. SAPIUM 2 0 2 0 0 0 0.3 0 0

FabaceaeFabaceae ‘4 looped 2aries’ FABA42 3 0 1 1 1 0 0.1 0.1 0.1Fabaceae ‘drip-tip’ FABADT 3 2 6 4 2 0.8 1 0.6 0.3Fabaceae ‘shapely’ FABASH 3 10 20 20 14 2.6 2 2 1.7Fabaceae ‘sloppy’ FABASL 3 0 2 0 0 0 0.3 0 0Fabaceae sp. FABACEAE 2 20 54 45 41 8.2 7.1 7.4 6.5

Fabaceae: CaesalpiniodaeBauhinia glabra Jacq. BAUHGL 3 5 18 13 12 2.1 2.6 2 1.9Bauhinia sp. BAUHINIA 2 0 0 0 1 0 0 0 0.1Senna herzogii (Harms) H.S. Irwin &

Barneby SENNHE 3 1 1 0 0 0.4 0.1 0 0

Senna sp. SENNA 2 0 1 1 0 0 0.1 0.1 0Fabaceae:Mimosoidae

Acacia ‘narrow’ ACACIANA 3 1 5 3 3 0.4 0.8 0.5 0.5Acacia sp. ACACIA 2 0 2 7 4 0 0.3 1 0.6Acacia ‘sp. 4’ ACACIA4 3 1 11 4 1 0.4 1.3 0.6 0.1Acacia ‘tiny’ ACACIATI 3 0 2 2 1 0 0.1 0.1 0.1Inga acreana Harms INGAAC 3 0 13 13 15 0 0.8 1 0.6Inga edulis Martius INGAED 3 1 4 3 4 0.4 0.6 0.5 0.6Inga klugii Standl. ex J.F. Macbr. INGAKL 3 0 0 0 1 0 0 0 0.1Inga marginata Willd. INGAMA 3 4 5 4 4 0.8 0.5 0.5 0.5Inga sp. INGA 2 95 247 221 199 34.7 31.4 29.1 27.6Parkia nitida Miq. PARKNI 3 0 0 1 1 0 0 0.1 0.1Parkia sp. PARKIA 2 0 5 1 1 0 0.6 0.1 0.1Zygia latifolia (L.) Fawc. & Rendle ZYGILA 3 1 5 2 1 0.4 0.6 0.3 0.1Zygia macrophylla (Spruce ex Benth.) L.

Rico ZYGIMA 3 8 12 10 12 2.1 1.5 1.2 1.5

Zygia sp. ZYGIA 2 2 0 0 0 0.4 0 0 0Fabaceae: Faboidae

Bocoa alterna (Benth.) R.S.Cowan BOCOAL 3 0 1 1 1 0 0.1 0.1 0.1Dalbergia monetaria L. f. DALBMO 3 1 2 4 4 0.4 0.3 0.5 0.5Dipteryx micrantha Harms DIPTMI 3 1 1 2 1 0.4 0.1 0.3 0.1Erythrina sp. ERYTHRINA 2 0 0 1 0 0 0 0.1 0Lecointea amazonica Ducke LECOAM 3 0 4 3 4 0 0.6 0.5 0.6Machaerium ‘3-2’ MACH32 3 0 20 7 5 0 2.2 1 0.8

111



Ormosia sp. ORMOMA 2 2 8 12 9 0.8 1.3 1.5 1.3Pterocarpus rohrii Vahl PTERRO 3 4 4 3 2 1.3 0.5 0.5 0.3Swartzia myrtifolia Sm. SWARMY 3 1 0 0 0 0.4 0 0 0

FlacourtiaceaeCasearia fasciculata (Ruiz & Pav.) Sleumer CASEFA 3 12 30 21 18 5.2 4.8 3.4 2.9Casearia javitensis Kunth CASEJA 3 1 1 0 0 0.4 0.1 0 0Casearia obovalis Poepp. ex Griseb. CASEOB 3 2 7 4 3 0.8 1.2 0.6 0.5Casearia sp. CASEARIA 2 0 8 24 23 0 1.3 3.1 3.6Hasseltia floribunda Kunth HASSFL 3 0 0 0 1 0 0 0 0.1Mayna odorata Aubl. MAYNOD 3 0 2 3 2 0 0.3 0.5 0.3Xylosma sp. XYLOSMA 2 0 1 1 0 0 0.1 0.1 0

GesneriaceaeCodonanthe uleana Fritsch CODOUL 3 0 2 2 0 0 0.3 0.3 0

HippocrateaceaeAnthodon decussatum Ruiz & Pav. ANTHDE 3 1 1 1 1 0.4 0.1 0.1 0.1Cheiloclinium cognatum (Miers) A.C.Sm. CHEICO 3 2 4 5 5 0.8 0.6 0.6 0.6Pristimera tenuiflora (Mart. ex Peyr.) A.C.

Sm. PRISTE 3 20 111 81 66 6.9 13 10.4 8.5

Salacia sp. SALACIA 2 4 10 11 11 1.3 1.5 1.7 1.7Icacinaceae

Calatola ‘microcarpa’ CALAMI 3 4 6 5 5 1.3 1 0.8 0.8Calatola venezuelana Pittier CALAVE 3 9 18 12 14 3.9 2.7 1.9 2.4Leretia cordata Vell. LERECO 3 4 8 7 13 1.7 1.3 1.2 2.2

LauraceaeCaryodaphnopsis fosteri van der Werff CARYFO 3 1 6 7 7 0.4 1 1.2 1.2Endlicheria dysodantha (Ruiz & Pav.) Mez ENDLDY 3 0 2 2 1 0 0.3 0.3 0.1Lauraceae sp. LAURACEAE 1 11 50 34 43 4.3 6.4 5.3 6.4Nectandra longifolia (Ruiz & Pav.) Nees NECTLO 3 6 34 23 29 2.1 5 3.6 4.5Ocotea sp. OCOTRI 2 0 0 0 2 0 0 0 0.1

LoganiaceaeStrychnos asperula Sprague & Sandwith STRYAS 3 3 7 6 6 1.3 1.2 1 1Strychnos tarapotensis Sprague & Sandwith STRYTA 3 2 6 4 4 0.8 1 0.6 0.6

MalpighiaceaeHeteropterys orinocensis (Kunth) A. Juss. HETEOR 3 0 2 2 2 0 0.1 0.1 0.1Hiraea reclinata Jacq. HIRARE 3 6 19 15 13 2.1 2.6 2 2Hiraea sp. HIRAEA 2 1 3 2 1 0.4 0.5 0.3 0.1Malpighiaceae sp. MALPIGIACEAE 1 3 10 14 19 1.3 1.5 2 2.9Malpighiaceae ‘willow’ WILLOW 3 0 7 1 0 0 0.1 0.1 0Mascagnia sp. MASCAGNIA 2 0 1 1 2 0 0.1 0.1 0.3

112



Stigmaphyllon sp. STIGMAPHYLLON 2 0 1 1 1 0 0.1 0.1 0.1Melastomataceae

Melastomataceae sp. MELASTOMATACEAE 1 1 1 0 0 0.4 0.1 0 0Melastomataceae ‘morada’ MELAMO 2 0 1 1 1 0 0.1 0.1 0.1Miconia sp. MICONIA 2 1 1 1 1 0.4 0.1 0.1 0.1Miconia triplinervis Ruiz & Pavón MICOTR 3 2 4 2 3 0.4 0.3 0.1 0.3Mouriri myrtilloides Spruce ex Triana MOURMY 3 0 3 2 2 0 0.5 0.3 0.3Mouriri peruviana Morley MOURPE 3 0 1 1 1 0 0.1 0.1 0.1

MeliaceaeGuarea macrophylla Vahl GUARMA 3 2 11 13 14 0.8 1.9 2.2 2.4Guarea pterorhachis Harms GUARPT 3 0 1 1 0 0 0.1 0.1 0Guarea sp. GUAREA 2 19 32 41 31 6.9 4.5 5.5 4.3Trichilia pallida Swartz TRICPA 3 1 3 3 2 0.4 0.5 0.5 0.3Trichilia pleeana (A. Juss.) C. DC.* TRICPL 3 4 249 151 100 1.7 12.5 10.4 8.6Trichilia poeppigii C. DC. TRICPO 3 18 36 30 29 6.9 5.3 4.6 4.6Trichilia quadrijuga Kunth TRICQU 3 1 3 3 3 0.4 0.5 0.5 0.5Trichilia rubra C. DC. TRICRU 3 2 3 3 3 0.8 0.5 0.5 0.5Trichilia sp. TRICHILIA 2 23 60 61 58 8.6 7.6 7.9 7.9

MenispermaceaeAbuta grandifolia (Mart.) Sandwith ABUTGR 3 0 1 2 2 0 0.1 0.3 0.3Abuta pahnii (Mart.) Krukoff & Barneby ABUTPA 3 0 1 1 2 0 0.1 0.1 0.3Anomospermum grandifolium Eichler ANOMGR 3 1 4 3 4 0.4 0.6 0.5 0.6Borismene japurensis (Mart.) Barneby BORIJA 3 1 4 4 2 0.4 0.5 0.6 0.3Chondrodendron tomentosum Ruiz & Pav. CHONTO 3 17 34 30 28 5.6 4.6 4.1 3.8Menispermaceae ‘pale’ MENIPA 3 3 7 6 6 1.3 1.2 1 1Menispermaceae sp. MENISPERM 1 7 12 17 17 3 1.9 2.6 2.6Odontocarya arifolia Barneby ODONAR 3 0 1 0 0 0 0.1 0 0Sciadotenia toxifera Krukoff & A.C. Sm. SCIATO 3 7 8 5 9 3 1.3 0.8 1.3

MonimiaceaeMollinedia racemosa (Schltdl.) Tul. MOLLRA 3 0 0 0 1 0 0 0 0.1Mollinedia ‘RF 8323’ MOLL83 3 2 3 3 4 0.8 0.5 0.5 0.6Mollinedia sp. MOLLINEDIA 2 5 4 5 4 1.7 0.6 0.8 0.6Siparuna sp. SIPARUNA 2 0 0 0 1 0 0 0 0.1

MoraceaeBatocarpus amazonicus (Ducke) Fosberg BATOAM 3 7 10 11 9 3 1.5 1.9 1.5Brosimum alicastrum Swartz BROSAL 3 5 11 15 25 2.1 1.9 2.6 3.8Brosimum lactescens (S. Moore) C.C. Berg BROSLA 3 2 17 12 10 0.8 2.2 1.5 1.3Brosimum sp. BROSIMUM 2 0 1 1 16 0 0.1 0.1 2.7Clarisia biflora Ruiz & Pavón CLARBI 3 2 5 4 6 0.8 0.8 0.6 1Clarisia racemosa Ruiz & Pavón CLARRA 3 3 36 24 29 1.3 4.3 3.6 4.6

113



Ficus sp. FICUS 2 1 2 1 1 0.4 0.3 0.1 0.1Moraceae sp. MORACEAE 1 5 6 2 1 2.1 1 0.3 0.1Naucleopsis sp. NAUCLEOPSIS 2 5 8 8 9 1.7 1.2 1.2 1.3Perebea guianensis Aublet PEREGU 3 4 6 4 6 0.8 0.6 0.5 0.6Perebea humilis C.C. Berg PEREHU 3 16 4 1 2 3.9 0.6 0.1 0.3Perebea sp. PEREBEA 2 3 9 10 9 1.3 1.3 1.5 1.2Poulsenia armata (Miq.) Standl. POULAR 3 1 1 1 1 0.4 0.1 0.1 0.1Pseudolmedia laevis (Ruiz & Pav.) J.F. Macbr. PSEULA 3 27 209 160 122 6.5 13.3 11.6 10.2Sorocea pileata W.C. Burger SOROPI 3 25 109 90 71 10.4 12.1 11.1 9.3

MyristicaceaeIryanthera juruensis Warb. IRYAJU 3 3 5 5 6 1.3 0.8 0.8 0.8Otoba parvifolia (Markgr.) A.H. Gentry OTOBPA 3 13 31 28 30 5.6 4.6 4.1 4.5Virola calophylla (Spruce) Warb. VIROCA 3 1 3 6 10 0.4 0.5 1 1.7Virola duckei A.C. Sm. VIRODU 3 2 2 1 1 0.8 0.3 0.1 0.1Virola flexuosa A.C. Sm. VIROFL 3 2 2 3 3 0.8 0.3 0.5 0.5Virola mollissima (Poepp. ex A. DC.) Warb. VIROMO 3 0 0 0 1 0 0 0 0.1

MyrsinaceaeArdisia nigrovirens J.F. Macbr. ARDINI 3 6 15 15 18 2.1 2.4 2.4 2.6Ardisia sp. ARDISIA 2 0 1 0 1 0 0.1 0 0.1Myrsinaceae sp. MYRSINACEAE 1 0 2 2 2 0 0.3 0.3 0.3Stylogyne cauliflora (Mart. & Miq.) Mez STYLCA 3 1 3 3 1 0.4 0.5 0.5 0.1Stylogyne micrantha (Kunth) Mez STYLMI 3 11 28 27 28 3 2.7 2.7 2.9

MyrtaceaeCalyptranthes lanceolata O. Berg CALYLA 3 1 2 2 2 0.4 0.3 0.3 0.3Calyptranthes ‘painted’ CALYPA 3 0 0 1 1 0 0 0.1 0.1Eugenia sp. EUGENIA 2 0 4 4 3 0 0.3 0.3 0.3Myrtaceae sp. MYRTAC 1 10 20 18 16 4.3 2.9 2.6 2.4

NyctaginaceaeNeea boliviana Standl. NEEABO 3 0 2 3 3 0 0.3 0.5 0.5Neea chlorantha Heimerl NEEACL 3 2 2 2 2 0.8 0.3 0.3 0.3Neea ‘RF 11702’ NEEA11 3 10 27 22 20 2.1 2.4 2 2Neea sp. NEEA 2 7 20 18 14 3 2.7 2.9 2.2

OchnaceaeOuratea sp. OURATEA 2 0 4 4 4 0 0.6 0.6 0.6

OlacaceaeHeisteria acuminata (Humb. & Bonpl.) Engl. HEISAC 3 1 2 2 2 0.4 0.3 0.3 0.3Heisteria nitida Engl.* HEISNI 3 1 3 31 18 0.4 0.5 3.8 2.2Heisteria sp. HEISTERIA 2 0 1 1 2 0 0.1 0.1 0.3Minquartia guianensis Aublet MINQGU 3 1 2 1 1 0.4 0.3 0.1 0.1

114



OpiliaceaeAgonandra brasiliensis Miers ex Benth. &

Hook. f. AGONBR 3 2 3 1 1 0.8 0.5 0.1 0.1

PhytolaccaceaePetiveria alliacea L. PETIAL 3 1 2 1 1 0.4 0.3 0.1 0.1

PiperaceaePiper laevigatum Kunth PIPELA 3 5 9 9 13 2.1 1.5 1.5 2Piper reticulatum L. PIPERE 3 2 5 4 4 0.8 0.8 0.6 0.6Piper sp. PIPER 2 10 2 0 0 3.4 0.1 0 0

PolygonaceaeCoccoloba lehmannii Lindau COCCLE 3 0 0 0 1 0 0 0 0.1Coccoloba peruviana Lindau COCCPE 3 0 4 3 5 0 0.5 0.5 0.5Coccoloba sp. COCCOLOBA 2 1 1 1 1 0.4 0.1 0.1 0.1Triplaris poeppigiana Wedd. TRIPPO 3 0 0 0 1 0 0 0 0.1

QuiinaceaeQuiina macrophylla Tul. QUIIMA 3 1 1 1 1 0.4 0.1 0.1 0.1

RhamnaceaeGouania lupuloides (L.) Urban GOUALU 3 3 5 4 3 1.3 0.8 0.6 0.5Rhamnidium elaeocarpum Reissek RHAMEL 3 1 4 4 2 0.4 0.6 0.6 0.3Ziziphus cinnamomum Triana & Planch. ZIZICI 3 9 51 45 29 3 4.6 4.3 3.1

RosaceaePrunus vana J.F. Macbr. PRUNVA 3 0 0 1 0 0 0 0.1 0

RubiaceaeAlibertia sp. ALIBERTIA 2 1 3 2 3 0.4 0.5 0.3 0.5Faramea anisocalyx Poepp. & Endl. FARAAN 3 1 1 1 2 0.4 0.1 0.1 0.1Faramea multiflora A. Rich. ex DC. FARAMU 3 10 39 31 22 4.3 5 4.8 3.6Faramea occidentalis (L.) A. Rich. FARAOC 3 7 15 14 13 2.1 2 1.9 1.9Faramea sp. FARAMEA 2 15 3 1 1 3 0.5 0.1 0.1Genipa americana L. GENIPAM 3 0 0 2 1 0 0 0.3 0.1Ixora peruviana (Spruce ex K. Schum.)

Standl. IXORPE 3 6 6 6 6 0.4 0.1 0.1 0.1

Posoqueria latifolia (Rudge) Roemer & Schultes POSOLA 3 1 4 5 5 0.4 0.6 0.8 0.8

Psychotria albert-smithii Standl. PSYCAL 3 1 1 1 1 0.4 0.1 0.1 0.1Psychotria bangii Rusby PSYCBA 3 1 0 0 0 0.4 0 0 0Psychotria sp. PSYCHOTRIA 2 1 1 1 2 0.4 0.1 0.1 0.3Psychotria viridis Ruiz & Pav. PSYCVI 3 2 3 4 6 0.4 0.3 0.5 0.8Randia armata (Swartz) DC. RANDAR 3 2 5 5 5 0.8 0.8 0.8 0.8Rubiaceae sp. RUBIACEAE 1 8 6 5 6 3 0.8 0.8 1Rudgea sp. RUDGEA 2 1 1 1 1 0.4 0.1 0.1 0.1

115



RutaceaeZanthoxylum sp. ZANTHOXYLUM 2 1 1 1 1 0.4 0.1 0.1 0.1

SapindaceaeAllophylus amazonicus (Mart.) Radlk. ALLOAM 3 3 5 4 5 1.3 0.8 0.6 0.8Allophylus divaricatus Radlk. ALLODI 3 1 2 4 9 0.4 0.3 0.6 1.5Allophylus glabratus Radlk. ALLOGL 3 3 7 6 5 1.3 0.8 0.8 0.8Allophylus sp. ALLOPHYLUS 2 0 25 8 4 0 2.9 1 0.6Matayba ‘M McFarland 971’ MATY97 3 16 58 66 43 5.6 6.5 6.4 5.3Matayba purgans (Poepp.) Radlk. MATYPU 3 0 2 1 1 0 0.3 0.1 0.1Paullinia ‘5 leaflets’ PAUL5L 3 1 1 1 1 0.4 0.1 0.1 0.1Paullinia ‘5 narrow lflts’ PAUL5N 3 2 4 4 4 0.8 0.6 0.6 0.6Paullinia alata G. Don PAULAL 3 0 1 1 1 0 0.1 0.1 0.1Paullinia bidentata Radlk. PAULBI 3 0 1 1 1 0 0.1 0.1 0.1Paullinia glomerulosa Radlk. PAULGL 3 0 3 2 1 0 0.5 0.3 0.1Paullinia hystrix Radlk. PAULHY 3 0 1 1 1 0 0.1 0.1 0.1Paullinia ‘leafy stipules’ PAULLLS 3 1 1 1 1 0.4 0.1 0.1 0.1Paullinia longifolia Radlk. PAULLO 3 0 1 1 1 0 0.1 0.1 0.1Paullinia mazanensis J.F. Macbr. PAULMA 3 0 1 1 1 0 0.1 0.1 0.1Paullinia ‘parallel 3 veined’ PAUL3V 3 1 5 4 4 0.4 0.8 0.6 0.6Paullinia reticulata Radlk. PAULRE 3 0 1 1 1 0 0.1 0.1 0.1Paullinia ‘rough three’ PAULR3 3 1 2 2 2 0.4 0.3 0.3 0.3Paullinia sp. PAULLINIA 2 26 64 63 64 9.1 8.8 9 8.5Paullinia tenera Poepp. PAULTE 3 4 7 6 6 1.7 1 0.8 0.8Pseudima frutescens (Aubl.) Radlk. PSEUFR 3 1 1 1 1 0.4 0.1 0.1 0.1Serjania dumicola Radlk. SERJDU 3 10 16 15 14 3 1.9 1.7 1.7Talisia peruviana Standl. TALIPE 3 48 89 82 77 8.6 6.7 6.4 6.5Toulicia reticulata Radlk. TOULRE 3 0 1 1 1 0 0.1 0.1 0.1

SapotaceaeMicropholis sp. MICROPHOLIS 2 0 0 1 1 0 0 0.1 0.1Pouteria caimito (Ruiz & Pav.) Radlk. POUTCA 3 0 3 3 3 0 0.3 0.3 0.3Pouteria ephedrantha (A.C. Sm.) T.D. Penn. POUTEP 3 8 17 18 19 3.4 2.7 3.1 3.1Pouteria juruana K. Krause POUTJU 3 0 0 0 1 0 0 0 0.1Pouteria procera (Mart.) T.D. Penn. POUTPR 3 3 6 6 6 1.3 1 1 1Pouteria sp. POUTERIA 2 8 13 16 15 3.4 2.2 2.7 2.6Pouteria torta (Mart.) Radlk. POUTTO 3 1 2 2 2 0.4 0.3 0.3 0.3Sapotaceae sp. SAPOTACEAE 1 2 4 1 1 0.8 0.6 0.1 0.1Sapotaceae ‘sp. 1’ SAPOTAC1 3 4 29 26 24 1.3 3.8 3.4 3.4Sarcaulus brasiliensis (A.DC.) Eyma SARCBR 3 1 4 4 4 0.4 0.3 0.3 0.3

SmilaceaeSmilax sp. SMILAX 2 0 4 3 2 0 0.6 0.5 0.3

116



SolanaceaeCuatresia fosteriana Hunz. CUATFO 3 0 1 1 1 0 0.1 0.1 0.1Solanaceae sp. SOLANACEAE 1 2 2 2 2 0.4 0.1 0.1 0.1Solanum acuminatum Ruiz & Pav. SOLAAC 3 0 0 0 1 0 0 0 0.1Solanum barbeyanum Huber SOLAVI 3 0 1 0 0 0 0.1 0 0Solanum sp. SOLANUM 2 2 2 2 4 0.4 0.1 0.1 0.5

StaphyleaceaeHuertea glandulosa Ruiz & Pav. HUERGL 3 4 7 9 5 1.7 1 1.2 0.5

SterculiaceaeByttneria asterotricha Mildbr. BYTTAS 3 4 11 9 9 1.7 1.5 1.2 1.2Sterculia sp. STERCULIA 2 0 0 0 1 0 0 0 0.1Theobroma cacao L. THEOCA 3 9 16 10 17 2.6 2 1.2 2.2

TheophrastaceaeClavija elliptica Mez CLAVEL 3 15 24 21 21 5.6 3.8 3.2 3.4Clavija sp. CLAVIJA 2 1 1 0 0 0.4 0.1 0 0Clavija tarapotana Mez CLAVTA 3 0 1 1 1 0 0.1 0.1 0.1

UlmaceaeAmpelocera ruizii Klotzsch AMPERU 3 0 9 5 6 0 1.2 0.6 0.8Celtis iguanaea (Jacq.) Sarg. CELTIG 3 3 23 3 2 1.3 3.1 0.5 0.3Celtis schippii Standl. CELTSC 3 1 9 8 7 0.4 1.5 1.3 1.2

UnknownUnknown unknown UNK 0 113 289 165 112 34.7 30.7 20.3 15.4

UrticaceaeUrera caracasana (Jacq.) Gaudich. ex Griseb. URERCA 3 0 1 2 0 0 0.1 0.3 0

VerbenaceaeAegiphila haughtii Moldenke AEGIHA 3 1 1 1 1 0.4 0.1 0.1 0.1Petrea aspera Turcz. PETRAS 3 0 1 2 2 0 0.1 0.3 0.3

ViolaceaeLeonia glycycarpa Ruiz & Pav. LEONGL 3 4 15 16 16 1.7 2.2 2.4 2.4Rinorea viridifolia Rusby RINOVI 3 9 35 32 32 3 5.2 4.6 4.6

VitaceaeCissus sp. CISSUS 2 1 0 0 0 0.4 0 0 0

Stem density of Focal species: 34 700 710 480Stem density of Non-Focal Species: 1229 3538 2902 2707

TOTAL: 1263 4238 3612 3187

Notes: “*” indicates the eight focal species used in seed-addition experiments of Chapter 2. Species names in single quotes denote morphospecies. Morpho-species names of the form ‘RF 11702’ indicate affinity with species collected by Robin Foster and deposited in the Field Museum herbarium (F). Taxonomy

117

primarily follows the Missouri Botanical Garden’s VAST (VAScular Tropicos) database (http://mobot.mobot.org/W3T/Search/vast.html). The column labelled

“C” indicates the degree of confidence in identification: 3: identification con-fident to species (or morphospecies), 2: genus level, 1: family level, 0: no confident identification. “Number of stems” indicated the total stem density of each species observed in each year. “Percent of plots” indicates the percentage of sample plots in which each species was observed. In 2003, 230 1-m2 plots were censused. In subsequent years, 572 plots were censused.

http://mobot.mobot.org/W3T/Search/vast.html

118

APPENDIX 2 – LETTERS OF PERMISSION

May 8, 2007

Sección de Permisos – Ecosistemas

Dr. Jorge Cortina

[email protected]

C. E. Timothy Paine

Dept. of Biological Sciences

Louisiana State University

Baton Rouge, LA 70803

225-578-7567

[email protected]

Querido Dr. Jorge Cortina:

Estoy cursando estudios graduados en la Universidad Estatal de Louisiana (LSU), EEUU.

Con mi asesor, Dr. Kyle E. Harms, publicó un articulo en �� que quisiera

incluir como un capitulo en mi disertación. Porque �� mantiene el derecho de

propiedad literaria, soy obligado por la universidad obtener permiso incluir este capitulo

en mi disertación. Pues, busco permiso re-imprimir un articulo que salió en ��

como un capitulo de mi disertación. Aquí se encuentra toda la información requisita por

�� para procesar esta petición.

Titulo: Regeneración de árboles tropicales e implicaciones para el manejo de

bosques naturales

Autores: Kyle E. Harms y C. E. Timothy Paine

Edicion: Año XII, Nº3 / 2003 Septiembre - Diciembre

Necesito incluir todo el texto. Voy a editar ligeramente el texto para conformarme con lapautas requeridas de disertaciones de la universidad. La sección de literatura citada serácombinada con la de los otros capítulos, según requisitos de la universidad. El artículoseguirá siendo de otra manera inalterado. Una nota al pie de la página será insertada alprincipio de capitulo reconocer el permiso de �� para utilizar el artículo. El artículoaparecerá como el primer capítulo de mi disertación. Seré el autor único de la disertación,que será sometida a la universidad como disertación electrónica no más adelante que el 1 dejunio de 2007. La disertación electrónica será archivada por la Biblioteca Middleton deLSU, en la biblioteca digital nacional de disertaciones, y con Proquest®. La bibliotecaimprimirá una sola copia que será archivado, desatado, en la biblioteca Hill de LSU. Ladisertación será disponible para todos los usuarios de estos sistemas.

Trataré puntualmente cualquier pregunta que pueda presentarse. Gracias por su ayuda coneste asunto.

Cordialmente,

C. E. Timothy Paine

119

From: [email protected]: Re: Permiso de imprimir un articuloDate: May 17, 2007 8:33:42 PM CDTTo: [email protected]

Apreciado Sr. Paine. Como editor de la revista Ecosistemas le comunico que tiene usted nuestra autorizacion para re-imprimir su articulo

Titulo: Regeneración de árboles tropicales e implicaciones para el manejo de bosques naturalesAutores: Kyle E. Harms y C. E. Timothy PainePublicacion: ECOSISTEMAS Año XII, Nº3 / 2003 Septiembre - Diciembre

en los terminos que menciona su carta, con el requisito de que sea mencionada la publicacion de origen del texto, tal como aparece en el parrafo anterior.

Deseandole suerte en su trabajo, reciba un cordial saludo,

Jordi CortinaEditorRevista Ecosistemas

From: [email protected] Subject: Permiso de imprimir un articulo Date: May 8, 2007 11:56:21 AM CDT

To: [email protected] Querido Dr. Cortina: Por favor, encuentra incluida con este una carta pidiendo permiso re-imprimir mi articulo, que salió en Ecosistemas, como un capitulo en mi disertación.

Cordialmente, C. E. Timothy PAINE

120

May 8, 2007

�� Permissions Department

[email protected]

C. E. Timothy Paine

Dept. of Biological Sciences

Louisiana State University

Baton Rouge, LA 70803

225-578-7567

[email protected]

Dear �� editor:

I am a PhD. candidate at Louisiana State University. With a collaborator, I published a

chapter of my dissertation has been accepted for publication in ��. Because ��

holds the copyright, I am required by the university to obtain permission to include this

chapter in my final dissertation. Thus, I seek permission to reprint an article that appeared

in �� as a chapter of my dissertation. Below is the information requested by

�� to process this request.

Title: SEED PREDATION BY NEOTROPICAL RAINFOREST MAMMALSINCREASES DIVERSITY IN SEEDLING RECRUITMENT

Authors: C. E. Timothy Paine and Harald Beck.

MS #: 06-1835

I need to include the entire text and all figures. I will somewhat rearrange the article to

conform to the dissertation consistency guidelines required by the university. The

literature cited will be combined with that of the other chapters, as per university

requirements. The article will otherwise remain unaltered. A footnote is to be inserted at

the start of the dissertation chapter acknowledging permission from �� to use the

article.

The article will appear as the fourth chapter of my dissertation, the title of

which will remain the same as the printed �� article. I will be the sole author

of the dissertation, which will be submitted to the university as an electronic

dissertation no later than 1 June 2007. The electronic dissertation will be archived

by Middleton Library of LSU, in the national digital library of theses and

dissertations, and with Proquest®. The library will print a single copy that will be

archived, unbound, in the Hill Memorial Library of LSU. The dissertation will be

available to all users of these systems.

I will promptly address any issues that may arise. Thank you for your

assistance with this matter.

Sincerely,

C. E. Timothy Paine

121

From: [email protected]: RE: 06-1835 permissionDate: May 14, 2007 7:03:51 AM CDTTo: [email protected]. Paine,

ESA policy gives you permission to reprint your article in your dissertation, as requested in the attachment to your email, and this email confirms that

permission.

Regards,

Clifford S. Duke

ESA Permissions Editor

From: C. E. Timothy Paine [mailto:[email protected]] Sent: Tuesday, May 08, 2007 12:24 PMTo: Cliff DukeSubject: 06-1835 permission Dear Ecology Editor: Please find attached a letter asking for permission to reprint my article, which has been accepted for publication by Ecology, as a chapter in my dissertation. Thank you-Tim Paine

122

VITA

Charles Eliot Timothy Paine was born in September 9, 1976, in Mentor, Ohio. Inspired by his

high-school biology teachers, Pat Barnes and Terry Harmon, he attended Dartmouth College and

majored in environmental biology and earth sciences. He graduated in 1999, having written a

senior honors thesis entitled “Holocene water quality and climate fluctuations in the New World

District, Montana”, under the guidance of Dr. C. Page Chamberlain. During and after college,

Tim had the opportunity to travel widely through the tropics. Most influential was the extensive

period he spent at Estación Biológica Cocha Cashu, in a remote, wild corner of the Peruvian

Amazon. There, his fascination with tropical diversity grew, and he developed interest in seedling

recruitment. In 2002 he began his doctoral studies in community ecology with Dr. Kyle Harms

at Louisiana State University. Tim is passionate about the diversity of the natural world, and the

processes by which it is maintained.

ecological factors affecting the diversity of …

Documents