UNIVERSITY OF HAWAI'I LIBRARY

SEXUAL REPRODUCTION AND THE EARLY LIFE HISTORY OF MONTIPORA

CAPITATA IN KANE'OHE BAY, 0'AHU, HAWAI'I

A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAI'I IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

IN

ZOOLOGY

MAY 2004

By Steven P. Kolinski

Dissertation Committee:

PaulL. Jokiel, Chairperson Robert A. Kinzie, III Evelyn F. Cox James E. Maragos Michael P. Hamnett © Copyright 2004 by Steven P. Kolinski

ll1 This work is dedicated to the memory ofmy father,

Frank Joseph Kolinski, Ph.D. (1931 - 1990),

and to MaryAnn,

Stella, and Jessica Ann Kolinski

IV ACKNOWLEDGEMENTS

I am grateful to Paul Jokiel, my committee chairperson, who opened his heart, mind and home and guided me towards research in coral reefecology, and Evelyn Cox, whose shared interests, insights, assistance and guidance helped to make this effort possible. I am also indebted to the other members ofmy committee, Bob Kinzie, Jim

Maragos and Mike Hamnett for insights, assistance and editorial review.

Various UH faculty are thanked for their support including Frank Stanton for advice and late night lab and field assistance, Andrew Taylor for statistical training and advice, Mike Hadfield for insights on larval biology, Dave Krupp for training and access to calorimetry equipment, and Ned Ruby for advice related to antibiotic treatments for larval settlement preference experiments. Cindy Hunter was instrumental in teaching me how to identify gametes in coral fragments, and Lloyd Watarai was invaluable in overcoming equipment challenges and assisted in the field.

Many student comrades and volunteers helped in various aspects ofthe field and lab work. In particular, thanks goes to Eric Brown, Ku'ulei Rogers, Usa Kuffner, Will

Smith, Franklin Te, Ann Tarrant, Shana Dukette, Erica Muse, Kanako Uchino, Eric

Conklin, Mark Mohlmann, Joseph Fal, Heather and Brian Reed, the 1997 Summer

Program participants, various East-West Center volunteers and many others.

Most ofall, I thank my family for their sustained patience and support throughout this effort. Parts ofthis research were supported by financial contributions from the East­

West Center and the Edwin W. Pauley Foundation (1997 Summer Program in Marine

Biology).

v ABSTRACT

The larval biology, early life history and energetic costs ofsexual reproduction in the scleractinian coral Montipora capitata were assessed in Kane'ohe Bay, O'ahu,

Hawai'i. Larvae were competent to settle three days post-spawning (DPS), and settled in large pulses days after syngamy in the laboratory and field. Ability to settle remained high (83 % proportional settlement) at six weeks, but was reduced (:::;;11 %) at 56 plus days. A few larvae survived and settled at 191 and 207 DPS, the longest lengths of competency recorded for a scleractinian coral. The larvae settled equally well on a variety ofsubstrates including different species ofcrustose coralline algae and rubble covered with filamentous algae. Tests with antibiotics suggested bacteria playa role in inducing settlement.

The early life histories ofMontipora capitata (hermaphroditic broadcast spawner), Porites compressa (gonochoric broadcast spawner) and Pocillopora damicornis (brooder) were monitored on field settlement plates. Montipora capitata had the highest number ofoverall settlers (91 % ofnearly 20,000 settlers), but displayed significantly lower levels ofsurvivorship and growth than P. compressa and P. damicornis. Each species displayed Type III survivorship. The mean estimated time for settlers to reach 1 cm2 projected area in the field was 4.9 years for M. capitata and 1.7 years for P. compressa and P. damicornis. Survival was not related to growth in M capitata, in contrast to the other two species. These three species differ in their fertilization and early life history strategies. Montipora capitata emphasizes processes that achieve large quantities oflarvae and settlers, while P. compressa and P. damicornis appear more dependent on settler quality (survival and growth).

VI The mean percent ofannual net photosynthetic productivity allocated to sexual

reproduction in Montipora capitata was estimated as 2.13 % for shallow reef-flat

colonies and 3.50 % for deeper reef-slope colonies. This energy expenditure could be

replaced by excess colony photosynthesis within two to six weeks (depending on depth)

following summertime spawning. Montipora capitata may gain its early life history

advantage through large repetitive, low cost, influxes oflarvae settling in a variety of habitats. Management implications are discussed.

VB TABLE OF CONTENTS Page

ACKNOWLEDGEMENTS...... v

ABSTRACT...... VI

LIST OF TABLES...... Xl

LIST OF FIG1JRES xiii

CHAPTER 1. INTRODUCTION...... 1

CHAPTER 2. LARVAL COMPETENCY AND SETTLEMENT PREFERENCE...... 6

Introduction...... 6 Methods 10 Field Studies...... 10 Culturing ofLarvae 13 Settlement Preference...... 14 Tests ofGeneral Substrate Preference.. 14 Antibiotic Treatments...... 16 GABA Treatments...... 19 Settlement Substrates in the Field...... 19 Settlement and Field Plate Conditioning...... 20 Settlement and Surface Orientation in the Field...... 21 Settlement Competency. 21 Proportional Settlement Over Time...... 21 Long-term Settlement Competency...... 23 Field Settlement Over Time...... 23 Results 24 Settlement Preference...... 24 Tests ofGeneral Substrate Preference 24 Antibiotic Treatments...... 25 GABA Treatments 27 Settlement Substrates in the Field...... 28 Settlement and Field Plate Conditioning...... 30 Settlement and Surface Orientation in the Field...... 30 Settlement Competency 30 Proportional Settlement Over Time...... 30 Long-term Settlement Competency 32 Field Settlement Over Time...... 32 Discussion...... 34 Settlement Preference...... 34 Settlement Competency...... 41

Vlll CHAPTER 3. SETTLEMENT, SURVIVAL AND EARLY GROWTH RELATIVE TO OTHER COMMON HAWAIIAN SCLERACTINIANS 47

Introduction " 47 Methods 49 Settlement Plates 49 Settlement Among Sites...... 50 Settlement Over Time 51 Settler SurvivaL...... 52 Montipora capitata 52 Comparison with Other Species..... 52 Settler Growth...... 54 Habitat Characteristics 55 Water Quality...... 55 Benthic Cover...... 57 Site Projection Example...... 57 Results 58 Settlement Among Sites...... 58 Settlement Over Time 60 Settler Survival...... 62 Montipora capitata...... 62 Comparison with Other Species...... 64 Settler Growth...... 65 Species Comparisons 65 Extended Single Polyp Status 67 Growth and Settler Survival...... 68 Settlement, Survival, Growth and Habitat...... 69 Site Projection Example...... 72 Discussion 74 Site and Seasonal Variation 74 Settler Survival...... 77 Settler Growth...... 79 A Comparison ofParameters and Suggested Recruitment Strategies 81

CHAPTER 4. ENERGY ALLOCATION TO REPRODUCTION...... 85

Introduction...... 85 Methods 87 Collection ofCorals and Measurement ofEnvironmental Parameters...... 87 Reproductive Output...... 90 Energy Estimates...... 91 Photosynthesis versus Irradiance Measurements...... 92 Estimates ofColony Productivity 94

IX Estimates ofAnnual Energy Allocation to Sexual Reproduction...... 95 Results 96 Environmental Parameters 96 Reproductive Output 97 Energy Content ofReproductive Products...... 98 Photosynthesis versus Irradiance Measurements and Estimates ofProductivity 99 Estimates ofAnnual Energy Allocation to Sexual Reproduction " 100 Discussion 102 Site Comparisons 102 Comparisons Between Species and Colony Processes 103 Limitations ofMethodology 106 Conclusions 108

SUMMARy 109

LITERATURE CITED 113

x LIST OF TABLES

2.1. Appearance, collection location and exposure ofcrustose coralline algae used in settlement preference experiments...... 15

2.2. Factorial ANOVA ofmean proportional M. capitata settlement on antibiotic treated versus FSW control substrates ,.. 25

2.3. Factorial ANOVA examining the effects oflarvae pre-exposed to antibiotics, substrate exposure to antibiotics and substrate type on mean proportional settlement ofM. capitata larvae 26

2.4. Factorial ANOVA ofthe proportion ofM capitata summer settlers on side, middle and bottom plate surfaces...... 31

2.5. Montipora capitata aquaria settlement on plates by census date under laboratory conditions...... 32

2.6. Settlement competency periods ofM capitata larvae under laboratory conditions...... 33

2.7. Natural settlement ofM capitata larvae observed over time at settlement survey sites in Kane'ohe Bay...... 33

2.8. Suggested classification ofscleractinian species based on available information ofobserved larval settlement preferences 38

3.1. Cumulative numbers ofscleractinian corals observed on settlement plates at six sites in Kane'ohe Bay from summer 1999 to summer 2001...... 59

3.2. Factorial ANOVA ofmean number ofM. capitata, P. compressa and P. damicornis settlers observed over two one-year monitoring periods.. 59

3.3. Mixed model ANOVAs examining temporal differences in mean estimated settlement m-2 ofcolonizable substrate at select sites in Kane'ohe Bay for individual scleractinian coral species 62

3.4. Species comparison ofproportionate survival at 4, 12 and 21 months...... 65

3.5. Species comparisons ofgrowth rates 67

3.6. Survivors noted as single polyps 12 months after first observation...... 68

Xl 3.7. a) Factorial ANOVA ofgrowth rates ofM. capitata, P. compressa and P. damicornis survivors (right censored individuals) and non-censored settlers; b) Binomiallogit homogeneity ofslopes model ofsurvival; c) Within species contrasts ofgrowth rates between right censored and non-censored settlers 69

3.8. Means and ranges ofrecorded habitat characteristics at monitored settlement sites in Kane'ohe Bay 70

3.9. Pearsons correlations matrix ofmeans ofmeasured habitat characteristics and mean number ofsettlers, settler growth and percent survival for three species across six sites in Kane'ohe Bay...... 71

3.10. Projected numbers ofsettlers and survivors in 500 mZ zones of measurement representing six sites in Kane'ohe Bay 73

3.11. Life history characteristics ofM capitata, P. compressa and P. damicornis in Kane'ohe Bay 82

4.1. Regression analysis ofcolony reproductive output (bundles) versus 3 colony size (cm ) '" 98

4.2. Factorial ANOVA ofenergy content (joules) per mg ofCoconut Island and Kokokahi gametes collected in 1997 and 1998...... 98

4.3. Factorial ANOVA ofmean bundle weights (mg) ofCoconut Island and Kokokahi gametes collected in 1997 and 1998...... 99

4.4. Photosynthesis versus irradiance hyperbolic tangent parameters, projected Oz production values and photosynthetic productivity estimates for Coconut Island and Kokokahi colonies between June 1996 and May 1997 101

XlI LIST OF FIGURES

Figure Page

2.1. Location ofsettlement survey sites in Kane'ohe Bay, O'ahu, Hawai'i...... 11

2.2. Settlement plate apparatus: a) Full apparatus; b) Plate set with seeding chamber 12

2.3. Mean proportional settlement ofM capitata larvae on various substrates... 24

2.4. Mean proportional settlement ofM capitata larvae on substrates treated with an antibiotic compared to substrates in FWS controls...... 25

2.5. Mean proportional settlement ofM. capitata larvae pre-exposed to antibiotic and FWS treatments in antibiotic and FSW substrate treatments '" 26

2.6. Mean proportional settlement ofM capitata larvae on substrates in antibiotic solution compared to substrates in FSW controls...... 27

2.7. Percent M capitata settlement on recorded substrates at settlement survey sites in Kane'ohe Bay...... 29

2.8. Proportion ofM capitata summer settlers on plate surfaces...... 31

3.1. LOglO mean number M capitata, P. compressa and P. damicornis settlers m-2 colonizable reefobserved over two one-year monitoring periods in Kane'ohe Bay...... 60

3.2. Temporal settlement ofselect species ofscleractinian corals in Kane'ohe Bay 61

3.3. Survivorship ofnewly settled M capitata larvae over time on three month conditioned plates seeded with larvae (1999) and three and 15.8 month conditioned plates with natural settlers (2000)...... 63

3.4. Mean proportional survivorship ofM capitata, P. compressa and P. damicornis settlers observed in Kane'ohe Bay 65

3.5. Mean growth rates ofM. capitata, P. compressa and P. damicornis 66

3.6. Mean growth rates ofM capitata, P. compressa and P. damicornis survivors (right censored individuals) and non-censored settlers 68

3.7. Generalized seawater circulation patterns in Kane'ohe Bay 76

Xlll 4.1. Coral collection sites in Kane'ohe Bay, O'ahu, Hawai'i 88

XIV CHAPTER 1

INTRODUCTION

Scleractinian corals are an essential component ofmost tropical and many subtropical oceanic reefsystems. They are generally responsible for much ofa reefs structural framework and habitat complexity and are an integral part ofmany elaborate reefbased communities. However, corals are susceptible to various natural and anthropogenic disturbances (Johannes 1975, Connell 1978, 1997, Brown and Howard

1985, Grigg and Dollar 1990, Maragos 1993, Jokie1 et al. 1993, Richmond 1993, Brown

1997, Karlson 1999, Hughes and Connell 1999, Pandolfi et al. 2003, Gardner et al. 2003), and appear particularly vulnerable to predicted changes in ocean temperatures, water chemistry, and sea level rises associated with climate change (Jokie1 and Coles 1990,

Smith and Buddemeier 1992, Brown 1997, Pittock 1999, Hoegh-Gu1dberg 1999, Hughes et al. 2003). The loss ofcorals from a community generally results in concomitant migration or death of associated reef fauna (Johannes 1975, Sano et al. 1984, Bell and

Ga1zin 1984, Rinkevich 1995, Connell 1997, see also Bruno and Bertness 2001). The dispersal, population dynamics and persistence ofcoral species are dependent on their strategies ofreproduction in relation to early life history success (Pearson 1981, Wallace

1985, Harrison and Wallace 1990, Lasker 1991, Veron 1995, Richmond 1997, Kinzie

1999, Ayre and Hughes 2000, Hughes and Tanner 2000).

Reproduction is one ofthe most critical components in the life histories ofall living organisms. It is a means by which species persist, adapt and in many cases disperse in spatially and temporally fluctuating environments (Williams 1975, Maynard Smith

1978, Bell 1982, Jablonski and Lutz 1983, Bull et al. 1987, Veron 1995, Holt and

1 McPeek 1996, Underwood and Keough 2001). Most corals reproduce both asexually and sexually. Common forms ofasexual reproduction include colony budding and fission

(Wells 1966, Hughes and Jackson 1985), fragmentation (Highsmith 1982), and for a few species, asexual production oflarvae (Stoddart 1983, Ayre and Resing 1986). A number ofstudies have highlighted the importance ofcolony fragmentation and fission to coral population dynamics (reviewed by Highsmith 1982, see also Wallace 1985, Hughes

1985, Jackson and Hughes 1985, Hunter 1993, Fong and Lirman 1995, Lasker 1991,

Lasker and Coffroth 1999). The dispersal potential ofthese processes is low, but it has been suggested their main effect is to amplify genotype survival and sexual reproductive success (Veron 1995, Coffroth and Lasker 1998, Lasker and Coffroth 1999). Sexual reproduction involves the formation oflarvae via syngamy. This results in genetically diverse progeny, which may enhance spatial variability in offspring recruitment

(Williams 1975, Maynard Smith 1978, Bell 1982). The larvae drift as meroplankton and provide, among other advantages, a means for range expansion and ability to colonize new habitats (Mayr 1970, Crisp 1974, Richmond 1987a, 1990, Grigg 1997, Pechenick

1999), and a means by which offspring can select habitats for recruitment (Fadlallah

1983, Morse et al. 1988, Harrison and Wallace 1990, Carlon and Olson 1993, Mundy and

Babcock 2000, Raimondi and Morse 2001, Negri et al. 2001).

About 25 years ago, investigators began focusing on contributions oflarvae to coral population recovery and maintenance (Connell 1973, Pearson 1981, Hughes and

Jackson 1985). Paradigms regarding larval recruitment's role in influencing coral community structure have shifted from the notion ofa taxonomically diverse oversupply oflarvae being acted upon by post-settlement selection factors to the confirmation of

2 taxonomic, spatial and temporal variability in larvae availability, settlement and recruitment success (Harrison and Wallace 1990, Fisk and Harriott 1990, Sammarco

1991, Tomascik 1991, Fitzhardinge 1993, Gleason 1996, Connell et al. 1997, Dustan and

Johnson 1998, Hughes and Tanner 2000, Hughes et al. 1999,2000,2002). Corals are a highly diverse group oforganisms with a variety ofreproductive strategies. Although many generalities are known, the specifics oflarval biology and early life history have only been elucidated for a small number ofcoral species (see Kinzie 1999, Nozawa and

Harrison 2003). Coral larval biology and early life history need to be understood across a range ofspecies and genera ifwe are to be able predict and potentially manage for coral responses to widespread environmental change (Kinzie 1999).

Reproductive patterns that lead to the production oflarvae can be classified for corals by sexuality and location ofembryogenesis (Harrison and Wallace 1990). Corals that form eggs and sperm within the same colony are hermaphroditic, while those that separate sexes among colonies are gonochoric. Species that develop and release planulae ofeither sexual or asexual origin are classified as brooders, while those that spawn their gametes into the surrounding water for fertilization and embryonic development are termed broadcast spawners. Nearly 85 % ofroughly 250 examined coral species broadcast spawn gametes (Richmond and Hunter 1990, Richmond 1997), and 75 % are reported to be hermaphrodites (Harrison and Wallace 1990, Richmond and Hunter 1990,

Richmond 1997, Veron 1995, 2000a). Broadcast spawning hermaphroditic species are found in at least 32 genera representing 9 families ofScleractinia (Richmond and Hunter

1990, Richmond 1997).

3 The general characteristics oflarval development in most species ofbroadcast spawning hermaphrodites appear to be similar (Harrison and Wallace 1990, Richmond

1997). A majority release positively buoyant eggs-sperm bundles that rise to the ocean surface for fertilization in an essentially two-dimensional environment. The eggs remain viable for up to seven to eight hours (Heyward and Babcock 1986), which increases the potential for fertilization success, even for species at low population densities. Larval development typically occurs within seven days (Babcock and Heyward 1986, Harrison and Wallace 1990, Richmond 1997), which, depending on water circulation provides initial opportunity for dispersal away from reefs oforigin. Larval competency, settlement, settler growth and survival are not well documented for the vast majority ofspecies within this group, and interspecific variation is likely. This dissertation provides information related to these aspects for the common Hawaiian hermatypic broadcast spawner, Montipora capitata.

The dispersal potential ofcoral larvae is dependent on their period ofcompetence to settle. Evidence suggests the larvae ofbroadcast spawning species can settle (i.e., attach to a substrate and metamorphose) rapidly (Harrison and Wallace 1990, Morse et al.

1996, Richmond 1997, Wilson and Harrison 1998, Nozawa and Harrison 2003, Miller and Mundy 2003). Yet, studies oflarval competency periods in broadcast spawning hermaphrodites are few (Richmond 1988, Wilson and Harrison 1998, Baird 1998,

Nozawa and Harrison 2003). In addition, evidence is accruing that coral larvae effect their distributions by actively selecting their settlement substrates, and settlement preferences do appear to differ between species (Fadlallah 1983, Morse et al. 1998,

Harrison and Wallace 1990, Carlon and Olson 1993, Mundy and Babcock 2000,

4 Raimondi and Morse 2000, Negri et al. 2001). Chapter 2 reports on laboratory and field experiments that elucidate larvae settlement preferences, inducers, settlement timing and periods ofcompetency in Montipora capitata.

Recruitment into a coral community is typically considered to occur once corals become easily visible in the field (Harrison and Wallace 1990). However, most coral larvae are small and early settlers ofmany species may only be visible when magnified

(Wallace 1983). Since the critical links between larval influxes and recruitment typically occur in the "invisible" (sensu Wallace 1983) realm, manipulative experiments must be utilized to identify and understand settlement and early life history patterns in corals.

Chapter 3 examines the settlement and long-term fate ofseeded and natural Montipora capitata larvae to artificial plates placed on shallow water reefs. Comparisons ofspatial and temporal settlement, settler growth and survival are made with a naturally settling brooding and gonochoric broadcast spawning species as a means to understand similarities and differences in early life history strategies.

Parental investments and costs associated with reproduction are not well known for coral species. However, resource allocation to reproduction is likely a reflection ofthe success oflarval and early life history strategy (Vance 1973, Begon et al. 1986, Morgan

1995). Chapter 4 examines energetic costs associated with sexual reproduction in

Montipora capitata with perspective to its early life history traits.

5 CHAPTER 2

LARVAL COMPETENCY AND SETTLEMENT PREFERENCE

Introduction

The planktonic larva is a life history characteristic common to scleractinian corals

(Harrison and Wallace 1990) and a majority ofbenthic marine invertebrate groups

(Thorson 1950, Mileikovsly 1971, Emlet 1991, Levin and Bridges 1995, Pechenik 1999).

This relatively small, free swimming, morphologically distinct reproductive phase has certain advantages for corals, including: a potentially low cost (Edmunds and Davies

1986, Morgan 1995, Jokie11998, Pechenik 1999, Kolinski Chapter 4), iterative and effective means ofproducing individuals that spreads cost and risk in relation to the likely generation time ofparental colonies; potential for spatial variability in offspring recruitment in a temporarily chaotic environment (Williams 1975, Jablonski and Lutz

1983, Holt and McPeek 1996, Pechenick 1999); reduced competition and inbreeding between parents and siblings and among siblings by means ofoffspring dispersal

(Strathmann 1974, Economou 1991, Knowlton and Jackson 1993, Pechenick 1999); reduced benthic predation on offspring during dispersal (Johannes 1978, Morgan 1995,

Pechenik 1999); range expansion and the ability to colonize new habitats (Mayr 1970,

Crisp 1974, Richmond 1987a, 1990, Grigg 1997); and a means by which offspring can select habitats for recruitment (Fadlallah 1983, Morse et al. 1988, Harrison and Wallace

1990, Carlon and Olson 1993, Mundy and Babcock 2000, Raimondi and Morse 2000,

Negri et al. 2001). The ubiquity and frequency oflarval production in Scleractinia underscores the ecological interconnectedness ofreefcorals within meta-populations

(Gaines and Lafferty 1995, Caley et al. 1996, Roberts 1997, Hubble 1997, Mumby 1999,

6 Hughes et al. 2000, Andrefouet et al. 2002), the boundaries ofwhich may be best defined by limitations to larvae dispersal and recruitment. Reefestablishment and connectivity by larvae may be critical to the persistence ofcorals over ecologic and geologic time

(Jablonski and Lutz 1983, Babcock 1988, Wallace 1985, Richmond 1990, Grigg and Hey

1992, Veron 1995, Jablonski 1995, Jackson 1995, Roberts 1997, Wilson and Harrison

1998, Kinzie 1999, Benzie 1999, Pechenick 1999, Vermeij 2001, Underwood and

Keough 2001). Connectivity is fundamental to the present day concept ofa "functional" marine reserve (Man et al. 1995, Roberts 1997, 1998,2000, Ogden 1997, Crowder et al.

2000, Chiappone and Sealey 2000, Dayton et al. 2000, Stockhausen et al. 2000, Palumbi

2001).

The effectiveness ofintraspecific aggregations ofcorals to act as seeding sources for local and/or regional colonization will depend not only on reproductive synchrony, high fertilization success and hydrodynamic processes, but also on the duration of embryonic and larval development in the plankton, the larvae settlement competency period, and on the overall settlement tendencies ofthe larvae. External development periods have been investigated for larvae ofa number ofcoral species (Kojis and Quinn

1981, Shlesinger and Loya 1985, Babcock and Heyward 1986, Richmond 1997, Mate et al. 1998, Schwartz et al. 1999), and the majority ofstudies find settlement within days of spawning or larval release (Duerden 1904, Abe 1937, Atoda 1953, Lewis 1974, Harrigan

1972, Richevich and Loya 1979, Kojis and Quinn 1981, Fadlallah 1983, Van Moorsel

1983, Shlesinger and Loya 1985, Szmant-Froelich et al. 1985, Richmond 1985a, 1997,

Babcock and Heyward 1986, Morse et al. 1988, Morse et al. 1996, Harrison and Wallace

1990, Johnson 1992, Carlon and Olson 1993, Wilson and Harrison 1998, Schwartz et al.

7 1999, Harii et al. 2001, Nozawa and Harrison 2003, Miller and Mundy 2003). However, few experiments have examined variability in coral settlement timing or attempted to elucidate overall periods ofsettlement competency (Van Moorsel1983, Richmond

1987a, 1988, Zaslow and Benayahu 1996, Wilson and Harrison 1998, Harii et al. 2001,

Nozawa and Harrison 2003). Many ofthe more general environmental factors influencing the settlement ofcoral larvae have been elucidated, including: the need for a hard substrate; species affiliations for surface rugosity (Harrigan 1972, Fadlallah 1983,

Carleton and Sammarco 1987, Harrison and Wallace 1990); the effects ofsediment accumulation (Hodgson 1990, Babcock and Davies 1991, Hunte and Wittenberg 1992, Te

1992, Gilmour 1999); and the influences oflight (Maida et al. 1994, Babcock and Mundy

1996, Mundy and Babcock 1998) and other physical factors (see Harrison and Wallace

1990, Richmond 1997). Yet, individual species settlement preferences are likely refined and may be quite discrete. Chemicals such as gamma aminobutyric acid, choline, serotonin, and 12-0-tetradecanoylphorbol-13-acetate induce settlement ofvarious marine invertebrate larvae under laboratory conditions (Morse et al. 1979, Heslinga and

Hillmann 1981, Rumrill and Cameron 1983, Pearce and Scheibling 1988, Nadeau et al.

1989, Morse 1990, Pawlik 1990, 1992, Henning et al. 1996, Avila et al. 1996, Fleck

1997, McCauley 1997). In addition, numerous studies have identified prokaryotic and/or eukaryotic organisms that induce settlement, most likely through chemical means

(Harrigan 1972, Morse et al. 1988, Morse 1992, Pawlik 1992, Carlon and Olson 1993,

Morse et al. 1996, Golbuu 1996, Leitz 1997, Johnson et al. 1997, Heyward and Negri

1999, Unabia and Hadfield 1999, Negri et al. 2001). Settlement preference experiments are now leading to the identification ofspecific chemical inducers in various types of

8 benthic biological communities that directly stimulate coral larvae settlement behavior

(Morse and Morse 1991, Morse et al. 1996). However, most settlement inducing communities (biological) and/or molecules remain unknown for the vast majority ofcoral

specIes.

The coral Montipora capitata (Anthozoa, Scleractinia; see Maragos 1977, 1995 for reported synonyms and reclassification) is widely distributed with geographic boundaries potentially reaching west Sumatra in the Indian Ocean, the Ryukyu Islands in the north Pacific, New Caledonia in the south Pacific and Hawai'i in the east (Veron

1993, 2000b; note the geographic range is in presently in dispute. Maragos 1995 suggests a range restricted to Johnston, the Hawaiian and Line islands). Although noted to be uncommon elsewhere (Veron 2000b), in Hawai'i it ranks third in statewide coral coverage in the Main Hawaiian Islands (Jokie1 et al. 2001) and Northwestern Hawaiian

Islands (Maragos et al. in press). Montipora capitata is found in a variety ofhabitats including deep, shallow, turbid, clear, calm and protected, and those directly exposed to high wave energy (Maragos 1972, 1977, Kolinski and JokieI2002). It is a broadcast spawning simultaneous hermaphrodite with buoyant egg-sperm bundles, surface water fertilization (Heyward 1986, Kolinski and Cox 2003) and an embryonic developmental period lasting two to three days (Mate et al. 1998, Kolinski pers. obs.). Very little is known, however, regarding its larval settlement tendencies or the period over which it remains competent to settle. This report presents results from a number oflaboratory and field experiments, elucidating larvae settlement preferences, inducers, timing and competency in this species.

9 Methods

Field Studies

Laboratory and field studies were conducted from 1998 through 2001 at the

Hawai'i Institute ofMarine Biology (HIMB) and in Kane'ohe Bay, O'ahu, Hawai'i.

Field observations ofMontipora capitata settlement were made on settlement plate sets

established at six shallow water (depths 1.5 to 3.5 meters) sites, including one fringing

(F) and patch (P) reefeach in north (N), central (C) and south (S) Kane'ohe Bay (Figure

2.1). Each plate set consisted oftwo 10.0 x 9.5 x 1.2 cm unglazed terra-cotta tiles

separated by a 0.75 x 1.5 x 2.5 cm piece ofPlexiglas and attached in a horizontal

position, two (1999-2000) to four (2000-2001) sets on each ofseven polyvinyl chloride

(PVC) posts at each site (Figure 2.2). Each set was held together with a 3.2 cm (1.25 in)

stainless steel bolt and a stainless steel washer and wing nut. Four 0.4 cm thick PVC clips

(1.9 cm diameter) were attached near each comer on each plate to allow for handling.

Posts were positioned every 5 to 10 meters along the dominant reefcontours, generally parallel to prevailing oncoming waves on the windward sides ofreefs, and were set between corals with bottom surfaces approximately 15 cm above sand, mud, rubble,

and/or hard reefsubstrates.

Initial plates (two to a post, seven posts at a site) were installed at sites in March, three months prior to the 1999 M capitata summer spawning, with replicate sets being

added three months prior to the 2000 summer spawning. Plate sets removed for analysis were always kept submerged in individual containers and were transferred by boat to the

Hawai'i Institute ofMarine Biology (HIMB) for settler identification, measurement and mapping using a Wild Heerbrugg dissecting microscope (with the aid ofa lOx 10 cm

10 157'45'

21"30'

o

N I o

21'25'

o 2 3 ======,===, km

157' 50' 157'45' Figure 2.1. Location ofsettlement survey sites in Kane'ohe Bay, O'ahu, Hawai'i (SF = South Fringing reef; SP = South Patch reef; CF = Central Fringing reef; CP = Central Patch reef; NF = North Fringing reef; NP = North Patch reef).

11 Figure 2.2. Settlement plate apparatus: a) Full apparatus; b) Plate set with seeding chamber. monofilament sectioned quadrat). In all cases, halfthe number ofsets on each post at a

site were collected at one time and analyzed, with collection schedules rotated among

sites to ensure equivalent time-related proportionate sampling. Plates were held in their individual containers in a shaded flow-through seawater table while in the laboratory.

12 Container seawater was filtered (40 /lm) and changed regularly. These procedures

ensured that plates were not contaminated with larvae in seawater at the lab. All plate sets were returned to their respective posts within one to two days. Examination ofall plate

sets at each census was conducted in less than one month.

In June 1999, roughly halfofthe plate sets (one from each offive to six posts at a

site) were inspected and then seeded with cultured M capitata larvae for related experiments. These sets were enclosed within individual transparent Glad brand plastic disposable containers (739 ml) modified with 40 /lm Nitex screen on all but the top surface to allow for adequate water exchange (Figure 2.2). The PVC clips ensured chamber (including mesh) and plate contact did not occur. Five hundred 6-day-old larvae were injected into each chamber in the field. Chambered sets were collected and settlement mapped and scored after 6 days. These seeded plates were returned (without chambers) to the field and were inspected with the other plates the following fall, winter­ spring and summer through summer 2001.

Culturing ofLarvae

Larvae ofMontipora capitata were raised under laboratory conditions. Colonies from various locations in Kane'ohe Bay were placed in flow-through seawater tables prior to the June, July and/or August new moon spawnings. Seawater flow to the tables was terminated one hour before spawning. Egg-sperm bundles from three or more colonies were collected from the water surface and mixed in translucent 500 ml

Nalgene™ polyethylene bottles (conditioned in seawater prior to use) filled one third full with either 0.45 /lm or 100 /lm filtered seawater (FSW). After two to three hours the containers and developing embryos were rinsed ofexcess sperm (using 0.45 /lm FSW

13 and 100-120 /lm Nitex screen). The embryos were returned to containers filled with 0.45

/lm FSW, which were then capped and placed in running seawater in shaded flow­ through seawater tables. Embryonic development proceeded fairly rapidly, with blastulae and gastrulae apparent 11 hours after spawning, round spinning ciliated larvae evident at

36 hours (1.5 days), and actively swimming barrel-shaped larvae prevalent at 60 hours

(2.5 days; see Mate et al. 1998). Bottles were cleaned ofdecomposed material daily by separating out live embryos and larvae using 100 - 120/lm Nitex screen, then scrubbing and rinsing the bottles with 0.45 /lm FSW. Over time, larvae were consolidated into fewer bottles. Water changes and scrubbing ofbottles containing larvae used for long­ term settlement competency experiments were accomplished every day post spawning for the first two to three weeks, but then proceeded on the order ofevery three to 21 days.

Settlement Preference

Tests ofGeneral Substrate Preference

Substrates for testing the settlement preference ofMontipora capitata larvae were collected from throughout Kane'ohe Bay. One branching and seven encrusting forms of crustose coralline algae (CCA) were collected from the tops, sides and undersides ofhard substrata. The eight CCA differed in appearance, collection location and exposure (Table

2.1), and were assumed to represent eight different species from an undetermined number ofgenera. Dead coral substrate covered with filamentous algae (FA), dead and dried coral rubble from Coconut Island (Moku 0 Loe) beach deposits, and terra-cotta tile fragments (from plates conditioned in a flow-through seawater table for 1.5 months) were also collected. Substrates were cut into 1 cm2 replicate samples, inspected for previous coral recruits, and were inspected and cleaned (except for the tile pieces) to ensure that

14 Table 2.1. Appearance, collection location and exposure ofcrustose coralline algae used in settlement preference experiments. Species Description of form and location CCAI Attached; branching; purple-pink; exposed to direct sunlight; reef flat Chinamen's Hat (Mo'okoli'i), North Bay. Identified as Hydrolithon setchelli (Shelly James pers. comm.). CCA2 Unattached; bulbous with large nodules; light purple; exposed to direct sunlight; reef flat Chinamen's Hat, North Bay. CCA3 Encrusting; rough with numerous conceptacles; red; cryptic, undersides of dead M. capitata plates; reef flat Wai'ahole-Waikane, North Bay. Identified as Hydrolithon gardineri (Shelly James pers. comm.). CCA4 Encrusting; smooth; pinkish-yellow; exposed to direct sunlight, covering shallow water carbonate; reef flat Marine Corps. Base sandbar reef, Central Bay. CCA5 Encrusting; rough; dark purple; cryptic, inside reef framework; reef flat Wai'ahole -Waikane, North Bay. CCA6 Encrusting; smooth and velvety; dark red; cryptic, inside reef framework; reef flat Coconut Island, Central Bay. Possibly Peysonella sp. Almost always observed competing with CCA8. CCA7 Encrusting; rough with numerous conceptacles, light red; exposed to direct sunlight; reef flat Coconut Island, Central Bay. Competing with adult M. capitata. CCA8 Encrusting, smooth; cryptic, inside reef framework; reef flat Coconut Island, Central Bay. Almost always observed competing with CCA6. Identified as Sporolithon c.f. ptychoides (Shelly James pers. comm.). desired biological forms were not contaminated by multiple algal or invertebrate species.

Each substrate sample was randomly assigned, along with controls (FSW only), to a cell within a set ofComing™ brand 6-cell, flat bottom culture trays (20 ml cells), with a total of5 replicates for each treatment. Trays were covered and translucent. All cells contained

0.45/.lmFSW.

Ten cultured, actively swimming, barrel-shaped larvae (5.5 days post fertilization) were pipetted into each treatment and control cell. All trays were partially submerged in flowing seawater in a semi-shaded flow-through seawater table to maintain seawater temperatures close to ambient. Every other day, approximately one third ofthe FSW in each cell was exchanged. On the seventh day, cells and substrates were examined at l2x using a Wild Heerbrugg dissecting microscope and scored for settlement. Larvae displaying strong adherence to a substrate and morphological change (from squat form to tentacle development) were considered settled. Proportional data were arcsine square-root

15 transfonned and analyzed using ANOVA. In the case ofno settlement in replicated FSW

controls (no substrate), control values were excluded from the ANOVA due to lack of

variance with the justification that proportional settlement in cells with substrates

represented the difference from mean settlement in controls. Post-hoc comparisons were

made using a Tukey HDS test with ex = 0.05.

Antibiotic Treatments

Experiment 1

A potential role ofbacteria in stimulating Montipora capitata larvae to settle was

examined by exposing substrates and larvae to an antibiotic treatment. Three species of

crustose coralline algae (CCA2, CCA6 and CCA8; those which induced the highest mean

settlement in the previous experiment) and dead coral substrate covered with filamentous

algae were collected from source areas within Kane'ohe Bay. Additionally, terra-cotta tile fragments were newly broken from settlement plates conditioned in running seawater

for 2.5 months in a flow-through seawater table. The substrates were trimmed to

approximately I cm2 in size, inspected for previous recruits, and were inspected and cleaned (except for the tile pieces) to ensure desired biological fonns were not

contaminated by multiple algal/other species. All substrate samples were randomly

assigned to control and experimental (antibiotic) groups with a total offive replicates for

each substrate for each treatment.

Experimental substrates were washed and soaked 5 times in separate 5 mg [1

concentrations ofantibiotic (chloramphenicol in 0.45 /..lm FSW) for 1 hr periods (total 5 hrs. each substrate). Control substrates were similarly washed and soaked in 0.45/..lm

FSW. Each substrate sample was placed in a randomly assigned cell in a set ofComing™

16 brand 6-cell culture trays with a medium corresponding to treatment (i.e. 0.45 !lm FSW or antibiotic solution). Trays were covered and translucent. Appropriate replicated control cells (antibiotic; no antibiotic) without substrates were established. Ten cultured, actively swimming, barrel-shaped M. capitata larvae (9 days post fertilization) were pipetted into each cell. The trays were partially submerged in running seawater in a semi-shaded flow­ through seawater table. Fluid mediums in each cell were replaced on a daily basis. On the seventh day cells and substrates were examined at 12x using a Wild Heerbrugg dissecting microscope and scored for settlement (defined above). Proportional data were arcsine square-root transformed for factorial ANOVA analysis. In the case ofno settlement in replicated FSW controls (no substrate), control values were excluded from the ANOVA due to lack ofvariance with the justification that proportional settlement in cells with substrates represented the difference from mean settlement in controls. Post-hoc comparisons were made using a Tukey HDS test with ex = 0.05.

Experiment 2

An additional experiment was conducted to determine whether the antibiotic might effect larval settlement through direct effects on the larvae as opposed to its effects on the substrate bacterial community. This experiment involved four treatment groups including: larvae pre-exposed to antibiotic put in cells with substrates containing antibiotic solution; larvae pre-exposed to antibiotic put in cells with substrates containing

FSW only; larvae pre-exposed only to FSW put in cells with substrates containing FSW only; and larvae pre-exposed only to FSW put in cells with substrates containing antibiotic solution.

17 Two species ofcrustose coralline algae (CCA6 and CCA8), dead coral substrate

covered with filamentous algae, and terra-cotta tile fragments newly broken from

settlement plates (conditioned in running seawater for 12 months in a flow-through

seawater table) were collected and processed as described above. Substrate samples were

randomly assigned to antibiotic and 0.45 /lm FSW treatment groups, and were separately

exposed to their respective treatments for a period ofsix days. The substrates were

maintained in preconditioned plastic cups halfsubmerged in running seawater in a semi­

shaded flow-through seawater table. Treatment fluids were exchanged daily and

substrates were rinsed rigorously.

Ten 500 ml Nalgene™ polyethylene bottles (translucent) containing high

concentrations ofcultured Montipora capitata larvae (7 days post fertilization) were

selected and randomly assigned, five to pre-exposure with antibiotic solution and five to

continued exposure to 0.45 /lm FSW only, for a period totaling seven days. Bottles were

capped and allowed to float in running seawater in a semi-shaded flow-through seawater table. Bottles were cleaned and fluids exchanged daily. Larvae and cleaning apparatus were exposed only to their respective treatment medium.

Each substrate sample was placed in a randomly assigned cell in a set of

Corning™ brand 6-cell culture trays with its corresponding treatment medium. The replicate bottles ofcultured larvae were combined by treatment, and ten actively

swimming, barrel shaped larvae were pipetted into each cell corresponding to their

assigned treatment. Antibiotic exposed larvae assigned to FSW cells were rinsed thoroughly and repeatedly with 0.45 /lm FSW prior to placement in their cells (to avoid contamination ofthe medium with antibiotic). All other groups oflarvae were treated

18 similarly within their respective treatment medium. Six replicates ofeach substrate in

each treatment were established along with replicated controls oflarvae in cells with only the treatment mediums. Trays were partially submerged in running seawater in a semi­

shaded flow-through seawater table to maintain seawater temperatures close to ambient.

Fluids were replaced on a daily basis. On the seventh day cells and substrates were

examined at 12x using a Wild Heerbrugg dissecting microscope and scored for settlement

(as defined above). Proportional data were arcsine square-root transformed for factorial

ANOVA analysis, with control (no substrate) values treated as described in Experiment

1. Post-hoc comparisons were made using a Tukey HDS test (a = 0.05).

GABA Treatments

The effect ofgamma aminobutyric acid (GABA) on Montipora capitata larvae settlement was examined by exposing replicates of 10 cultured, actively swimming, barrel-shaped larvae (13 days post fertilization) to approximately 10 ml of 10'6M, 10,sM,

10·4M, 10'3M, and 10·2M concentrations ofGABA (in 0.45 /lm FSW) in FSW conditioned glass scintillation vials (see Morse et al. 1979). Five replicates ofeach treatment were established with replicate controls (0.45 /lm FSW), and were capped and maintained halfsubmerged in running seawater in a semi-shaded flow-through seawater table. Vials were examined every day for three days and at three weeks at 12x with a

Wild Heerbrugg dissecting microscope and scored for settlement and survival.

Proportional data were arcsine square-root transformed prior to statistical analysis.

Settlement Substrates in the Field

Natural and artificial substrates selected for settlement were recorded and

19 categorized in an effort to identify the range ofsubstrates selected for natural and seeded

settlement on field plate sets over a two year period. A description ofthe data is

presented. Preference could not be determined statistically as data collection was

haphazard and substrate availability between sites was inconsistent. Substrates upon

which larvae settle do not necessarily identify inducer communities (Morse et al. 1988,

Morse et al. 1996, Heyward and Negri 1999, Negri et al. 2001). Available substrates with no settlement were also recorded.

Settlement and Field Plate Conditioning

Settlement preference for plate sets differing in lengths oftime offield conditioning was examined in two separate trials at each offour field sites. Summer

settlement in 2000 was compared between 3.3 and 15.8 month conditioned plates, and in

2001 between 15.8 and 28.3 month conditioned plates. Full examination ofplates occurred over a three-week period in a pair-wise fashion (i.e., one new and one old plate

set from each post at a site were always examined on the same day). Preference was examined by calculating the proportional difference between the sum ofall settlers on plates ofsimilar age on each post (two plates for each age group), divided by the total of all settlers on the four plates ofeach post ((sum ofold - sum ofnew)/total). Posts with less than two natural settlers were excluded from the analysis; thus, comparisons were restricted to Central and South Bay sites. Only natural settlement in summer months was scored. Data for each year were analyzed separately using ANOVA, testing whether the mean preference differed from zero (no preference) at the four sites. Post-hoc comparisons were made for each site in each year using single sample t-tests applying a

Bonferroni correction for multiple comparisons (Moore and McCabe 1993).

20 Settlement and Surface Orientation in the Field

Proportional settlement on plate surfaces was compared to ascertain larval settlement preference for surface orientation. Plate surfaces were categorized as top, bottom, middle and sides. Natural summer settlement ofMontipora capitata in each surface category was determined and pooled across plates within each offour sites, including South Fringing, South Patch, Central Fringing and Central Patch reefs in 2000 and 2001. Low settlement at North Bay sites prohibited their inclusion. Proportional data were arcsine square-root transformed. A Factorial ANOVA and Tukey HDS tests (a =

0.05) were used to examine surface preference, with the plate top surface category removed prior to analysis (due to negligible settlement) to meet model assumptions. Year was treated as a replicate factor, but was dropped from the final analysis due to limited degrees offreedom and its insignificant effect in exploratory analyses.

Settlement Competency

Proportional Settlement Over Time

Proportional settlement ofcultured Montipora capitata larvae over time was examined in five replicate nine-liter glass aquaria. Five 10.0 x 9.5 x 1.2 cm unglazed terra-cotta tile plates were conditioned in running seawater in a semi-shaded flow­ through seawater table for approximately two months. Prior to use in the experiment, these plates were examined to ensure cora11arvae caught in the seawater system had not settled to the plates during conditioning. Sea anemones that had settled on the plates were removed. Each plate was rinsed thoroughly with 0.45 /lm FSW and was placed, with six liters of0.45 /lm FSW, in an individual aquarium. Three 0.4 cm thick PVC clips (1.27 cm diameter, pre-conditioned in running seawater) were attached to each plate. This raised

21 the plates above aquaria glass bottom surfaces and allowed for plate transfer without directly handling plate surfaces. The aquaria were maintained halfsubmerged in running seawater in a semi-shaded flow-through seawater table to maintain seawater temperatures close to ambient. Daytime temperatures in the aquaria ranged between 26.5 and 27.9°C.

Water levels in each aquarium were maintained over time by adding 0.45 /..lm FSW.

Aeration and water motion were maintained in each aquarium by bubbling a slow but constant stream ofair through a Pasteur pipette fitted to an air-line.

One thousand cultured M capitata larvae (2.5 days post spawning) were added to each aquarium. The following day, each plate was transferred to a large glass bowl containing 0.45 /..lm FSW, briefly examined at 12x with a Wild Heerbrugg dissecting microscope, and returned. Plates consistently remained submerged during transport and inspection. Settlement was scored and locations mapped (with the aid ofa lOx 10 em monofilament sectioned quadrat) for each plate surface four, seven, 12, 17 and 22 days post spawning. Larvae displaying strong adherence to the substrate with evident morphological change (from squat form to tentacle development) were considered settled. The low levels ofsettlement observed on the clips and aquarium walls were recorded but not included in the plate settlement tallies. Although new settlement was observed to have occurred, significant mortality ofsettlers on two ofthe plates between days seven and 12 made their results on day 12 difficult to interpret. Original plates were replaced by similarly sized and conditioned terra-cotta tiles on day 12 to ease mapping efforts. A repeated measures ANOVA comparing settlement dai1 between each ofthe first three census periods (times within which settlement occurred) was conducted for the three plates not affected by mass mortality. Settlement values were 10glO transformed to

22 meet model assumptions. All aquaria were considered to be saturated with larvae; thus,

settlement rates were not proportionally adjusted (circumventing the need for multiple transformations on data that exhibited analogous trends).

Long-term Settlement Competency

Long-term settlement competency ofMontipora capitata larvae was investigated directly through experimentation. Actively swimming, barrel-shaped larvae raised in 500 ml Nalgene™ culture bottles were added to cells in Coming™ brand 6-cell culture trays containing 0.45 Ilm FSW with filamentous covered coral substrate, CCA6 and/or CCA8.

The trays were partially submerged in running seawater in a semi-shaded flow-through seawater table to maintain seawater temperatures close to ambient. Cells were monitored over time with FSW changes occurring daily, weekly or biweekly as needed. Settlement

(as defined above) was scored at each inspection, and competency was calculated as the interval between the date ofcolony spawning and relevant inspection dates flanking settlement events. Monitoring continued until all viable larvae had settled or perished.

Additional information related to larval settlement competency was gained through examination oflarval age and settlement timing in the larval settlement preference experiments.

Field Settlement Over Time

Natural Montipora capitata settlement on 33 to 168 recruitment plate sets established at the six Kane'ohe Bay field sites (Figure 2.1) was examined with respect to settlement timing as defined by six census periods over a two-year period (1999 to 2001).

Data from 11 plate sets were pooled to give a total proportional estimate ofnew settlement versus census period. Data corresponding to similar monitoring during the 1999 summer

23 spawning period were not included, as replicates were not equally examined over the course ofthe year due to the larval seeding experiments.

Results

Settlement Preference

Test ofGeneral Substrate Preference

Mean settlement (± S.E.) ofMontipora capitata larvae on tested substrates is shown in Figure 2.3. A significant difference was found in settlement between substrates

(ANOVA, n = 55, FlO, 54 = 3.12, P = 0.004). Two groups ofsubstrates overlapped in terms ofsettlement preference, but only CCA6 and CCA8 (encrusting forms ofcrustose coralline algae) experienced significantly higher levels ofsettlement than CCAI

(branching form ofcrustose coralline algae) and unconditioned beach rubble. No settlement was evident in the controls.

0.6

0.5

'"0 ~ 0.4 Q5 en c o 15 0.3 a. a..e ffi 0.2 a.> :2:

0.1

o

Substrate Figure 2.3. Mean (± S.E.) proportional settlement ofM. capitata larvae on various substrates (CCA = crustose coralline algae; FA = filamentous algae covered coral skeleton; * = no settlement).

24 Antibiotic Treatments

Experiment 1

No settlement oflarvae occurred in cells without substrates. Substrates exposed to the antibiotic treatment showed significantly lower levels ofsettlement across all substrate groups (Table 2.2, Figure 2.4). There were no significant differences in settlement between the three CCA types and the filamentous algae covered coral skeleton; however, the four differed significantly from plate fragments (Table 2;

FA=CCA8=CCA6=CCA2>Plate) in contrast to earlier findings (Figure 2.3).

Table 2.2. Factorial ANOVA ofmean proportional M capitata settlement (arcsine square-root transformed) on antibiotic treated versus FSW control substrates. Source df SS MS F p Treatment 1 1.7065 1.7065 20.73 0.0000 Substrate 4 4.8914 1.2229 14.86 0.0000 Treatment x Substrate 4 0.0086 0.0022 0.03 0.9986 Residual 40 3.2922 0.0823 Total 49 9.8987 1 • Control

0.9 ~ Antibiotic 0.8 "0 (J) :::::: 0.7 (J) (f) c 0.6 o ~ 0.5 D- e 0.4 CL.. ~ 0.3 (J) :2: 0.2 0.1 o FA CCA8 CCA6 CCA2 Plate Substrate Figure 2.4. Mean (± S.E.) proportional settlement ofM capitata larvae on substrates treated with an antibiotic compared to substrates in FSW controls (CCA = crustose coralline algae; FA = filamentous algae covered coral skeleton).

25 Experiment 2

Larvae did not settle in cells without substrates. No significant effect oflarvae pre-exposure to antibiotic was detected (Table 2.3). Pre-exposed larvae actually displayed a tendency for higher mean proportional settlement across treatments (Figure 2.5).

Substrate treatments with antibiotic solution had significantly lower settlement than comparable controls for all substrates with the exception ofCCA8, which displayed a similar trend (Table 2.3, Figure 2.6). There was a significant substrate effect for the combined treatments (Table 2.3; FA = CCA8 > CCA8 = CCA6 > CCA6 = Plate).

Table 2.3. Factorial ANOVA examining the effects oflarvae pre-exposure to antibiotics, substrate exposure to antibiotics and substrate type on mean proportional settlement ofM capitata larvae. Source df SS MS F p Larvae Pre-Trmt (L) 1 0.1455 0.1455 2.25 0.1377 Substrate Trmt (ST) 1 2.5524 2.5524 39.44 0.0000 Substrate (S) 3 1.4287 0.4762 7.36 0.0002 LxST 1 0.0095 0.0095 0.15 0.7033 LxS 3 0.0512 0.0171 0.26 0.8518 STxS 3 0.5356 0.1785 2.76 0.0469 Lx STx S 3 0.0150 0.0050 0.08 0.9667 Residual 80 5.1776 0.0647 Total 95 9.9155

• Larvae Pre-treated with Antibiotic 1 ~ Larvae Pre-treated with FSW 0.9 "00.8 Q.) :m 0.7 (f) c 0.6 o ~ 0.5 Cl.. eOA 0- ~ 0.3 Q.) ~ 0.2 0.1 o Control Antibiotic Substrate Treatment Figure 2.5. Mean (± S.E.) proportional settlement ofM. capitata larvae pre-exposed to antibiotic and FSW treatments in antibiotic and FSW substrate treatments (data for substrates within a treatment group combined).

26 • Con tro I

0.9 I§]] An tib io tic

0.8

-= 0.7 Q) ~ 0.6 c: o 'E 0.5 ec... a... 0.4 c: co Q) 20.3

0.2

0.1 o FA C C A 8 C C A 6 Pia te Sub s tra te Figure 2.6. Mean (± S.B.) proportional settlement ofM capitata larvae on substrates in antibiotic solution compared to substrates in FSW controls (CCA = crustose coralline algae; FA = filamentous algae covered coral skeleton).

GABA Treatments

No settlement or metamorphic responses (polyp balls) were observed in any treatment within three days following exposure to GABA. GABA at lO-2 M was toxic to

Montipora capitata larvae, with only one of50 larvae (2 %) surviving after two days exposure. GABA also appeared to be toxic at lO-3 M to larvae over a three-week period.

Qualitative behavioral responses appeared to differ between treatments within the first three days. Larvae in the 10-3 M treatment displayed little movement, those in the 10-4 M treatment were noticeably sluggish, while active swimming was observed in the 10-5 M,

10-6 M and control treatments. At the end ofthree weeks, all 10-2 M and 10-3 M exposed larvae were dead, two larvae each had settled and metamorphosed in the 10-4 M (mean

survival = 46 ± 12 % S.B., n = 5) and 10-6 M (mean survival = 76 ± 19 % S.E., n = 5)

5 treatments, no 10- M exposed larvae had settled (mean survival = 82 ± 6 % S.E., n = 5), and one control larvae had settled and metamorphosed (mean survival = 84 ± 13 % S.E.,

27 n = 5). No significant difference in settlement existed between 10-4 M GABA, 10-6 M

GABA and 0.45 /lm FSW control treatments (Kruskal-Wallis AOY, n = 15, H = 0.56, P

= 0.756). Survival was not affected by GABA presence at tested concentrations below

3 10- M (ANaYA, n = 20, F3, 19 = 1.61, P = 0.226).

Settlement Substrates in the Field

Settlement on plate sets was common on smooth "exposed" surfaces and in cryptic grooves and crevices ofthe tiles and encrusting communities. Settler surface distributions ranged from highly aggregated to well dispersed. Specific substrate types were recorded for 16 % ofthe recent settlements observed (n = 3,321; South Fringing =

390 settlers, South Patch = 163, Central Fringing = 1845, Central Patch = 711, North

Fringing = 71, North Patch = 141). Substrates receiving greater than one percent of settlements at a site are shown in Figure 2.7. Montipora capitata larvae settled on a variety ofencrusting community species as well as on conditioned tile surfaces. Yarious forms ofcrustose coralline algae hosted large percentages oflarval settlement at North and Central Kane'ohe Bay sites, but were not the dominant supporting substrates at

South Bay sites. Oyster shell surfaces created cryptic habitat and also supported large numbers ofsettlers, particularly in South and Central Kane'ohe Bay. Settlement on macroalgae (Dictyosphaeria cavernosa), solitary tunicates, dead coral skeleton, live and dead branching and live encrusting bryozoans, and slime mold surfaces was observed but was rare « 1 %). Common substrates not observed to host settlement included all observed sponges, colonial tunicates, live corals, sedentary soft tube polychaetes, hydroids, chitons, fish eggs, various macroalgae (including Caulerpa sp.and Bryopsis sp.) and sediment.

28 100.0 • SF Lill CP 90.0 80.0 D SP til NF ..... c 70.0 Q) E • CF ~ NP Q) 60.0 :;:;..... Q) 50.0 CJ)..... c Q) 40.0 () "- Q) 30.0 N a.. \0 20.0 10.0 0.0 Bk CCA DCCA DEB DM Oy Day Fil Ht Mix SI Vm Substrate Figure 2.7. Percent M capitata settlement on recorded substrates at settlement survey sites in Kane'ohe Bay (SF = South Bay Fringing reef; SP = South Bay Patch reef; CF = Central Bay Fringing reef; CP = Central Bay Patch reef; NF = North Bay Fringing reef; NP = North Bay Patch reef). Substrates include: Bk = blank tile with presumed biological film; CCA = crustose coralline algae; DCCA = dead CCA; DEB = dead encrusting bryozoan; DM = domed mollusk; Vm = vermetids; Oy = oyster; Doy = dead oyster shell; Fil = filamentous algae; Ht = Hydroides sp. tube, and; Mix = larval base in contact with more than one substrate type. Settlement and Field Plate Conditioning

The mean settlement preference index was -0.068 (± 0.119 S.B.) for the 3.3 and

15.8 month comparison and -0.040 (± 0.071S.E.) for the 15.8 and 28.3 month comparison. Neither index differed significantly from zero, and site indices did not differ significantly from each other, suggesting no statistically notable natural settlement preference for plates based on the levels ofconditioning (2000 comparison, ANOYA, n =

24, F3, 20 = 2.03, P = 0.142; 2001 comparison, ANOYA, n = 28, F3,24 = 1.30, P = 0.298).

Post-hoc t-tests ofindividual sites in each year did not find significant differences from zero (P > 0.05) when corrected for making multiple comparisons.

Settlement and Surface Orientation in the Field

Percent summer settlement to plate surfaces averaged 0.04 ± 0.03 % S.E. on plate tops; 25.01 ± 4.09 % S.E. on plate sides, 42.30 ± 9.91 % S.E. on inner surfaces, and

32.64 ± 7.26 % S.E. on plate bottoms. Overall, the proportion ofsettlers on top surfaces was lower than that on each ofthe other surfaces (Figure 2.8), and significantly higher levels ofsettlers where identified on inner surfaces than sides (Table 2.4). Settlement preference for all but the top surfaces varied by site (Figure 2.8, Table 2.4).

Settlement Competency

Proportional Settlement Over Time

Although not quantified, large numbers (estimated in the hundreds) oflarvae were observed attached to each ofthe plates three days following spawning. Settlement data at each census date are shown in Table 2.5. The mean number ofnew Montipora capitata settlers was highest in the initial census four days after spawning and declined

1 substantially with time. Settlement rates (settlers dai ) were significantly different over

30 Top 1 • WJfJ Sides 0.9 0 Middle 0.8 U) l1li Botto m ID 0.7 = ~ 0.6 ...... o 0.5 c o t 0.4 o g- 0.3 a...L- 0.2 0.1 a SF SP CF CP Site Figure 2.8. Proportion ofM capitata summer settlers on plate surfaces (S = South; C = Central; F = Fringing reef; P = Patch reef).

Table 2.4. Factorial ANOVA ofthe proportion ofM capitata summer settlers (arcsine square-root transformed) on side, middle and bottom plate surfaces at South and Central Fringing and Patch reefs. Source df SS MS F p Site 3 0.0007 0.0002 0.0163 0.9970 Surface 2 0.1348 0.0674 4.4812 0.0352 Site*Surface 6 0.9701 0.1617 10.7520 0.0003 Error 12 0.1805 0.0150

time, with rates at Time 1 > Time 2 > Time 3 (repeated measures ANOVA, n = 3, F2,4 =

131.77, P = 0.0002). Two larvae were observed attached to plate 2 on day 17, but could not be classified as settled.

The measured proportion of available larvae that cumulatively settled on all plates

averaged 45.1 % (± 3.1 % S.B., n = 5). An additional 12 settlers were noted on clips

(mean = 2.4 ± 2.2 S.B., n = 5) and 47 larvae were found settled on aquarium walls (mean

= 9.4 ± 2.2 S.B., n = 5). Upon termination ofthe experiment, free-swimming larvae were recovered but were few in number (mean = 4.2 ± 1.9 S.B., n = 5).

31 Table 2.5. Montipora capitata aquaria settlement on plates by census date (days after spawning) under laboratory conditions (* new settlement on two plates could not be accurately measured due to high mortality ofpreviously settled spat). No. Days Post Mean No. New Spawning Settlers (S.E.) Range Cumulative Range n 4 363.2 39.9 228 - 473 228 - 473 5 7 74.4 9.3 47 - 103 304 - 520 5 12 23.0 3.2 18 - 29 333 - 475 *3 17 0 ( - ) o 333 - 520+ 5 22 0 ( - ) o 333 - 520+ 5

Long-term Settlement Competency

Montipora capitata larvae settled in proportionately high numbers (83 %) at 42 days post spawning. However, settlement was reduced at 56 plus days (settlement ::;11

%; Table 2.6). Although few in number, some M capitata larvae displayed an ability to settle well beyond 120 days. Settlements at 191 to 197 and 207 to 213 days were monitored for an additional 109 days. These settlers metamorphosed into normal sized single polyps that survived throughout the monitoring period and maintained appearances consistent with those observed naturally in the field.

Field Settlement Over Time

Field settlement over time appeared to be highest within the spawning season (late

May to August with peaks in June and July; Table 2.7; see Kolinski Chapter 3). The percentage ofnew Montipora capitata settlers identified during July and August censuses totaled 98.02 % over the two-year period. A total 1.86 % ofnew settlers were noted in the

October censuses, and only 0.12 % ofnew settlers were noted in February and March in

2000 and March and April in 2001. Although these estimates cannot reliably be corrected for mortality ofsettlers not observed, consistent order ofmagnitude differences over time suggests proportionately high settlement soon after spawning, but there appears to be some potential for settlement delay.

32 Table 2.6. Settlement competency periods ofM capitata larvae under laboratory conditions (Note: two additional trials between 9-29-98 and 11-19-98 resulted in no settlement). Competency Spawning Settlement (days) Date Date Description 9 to 16 7-26-98 8-5-98 to 89% (179 of201) settlement. Ten larvae placed in each 8-12-98 oftwenty cells, five with filamentous algae covered coral skeleton, five with CCA2, five with CCA6 and five with CCA8. No settlement differences between substrates. No settlement in controls. Seven day exposure oflarvae to substrates.

42 to 45 7-26-98 9-7-98 to 83% (33 of40) settlement. Eight larvae placed in each of 9-10-98 five cells with filamentous algae covered coral skeleton. No settlement in controls. Seven day exposure to substrates.

56 to 64 7-26-98 9-21-98 to 11 % (1 of9) settlement. Three larvae placed in each of 9-29-98 three cells with CCA8. No settlement in controls. Eight day exposure to substrates.

145 to 164 6-26-98 11-19-98 to 7% (1 of 15) settlement. Five larvae placed in each of 12-8-98 three cells with filamentous algae covered coral skeleton and CCA6 and CCA8. No controls. Thirty-two day exposure to substrates.

207 to 213 6-26-98 1-20-99 to 7% (1 of 15) settlement. Five larvae placed in each of 1-26-99 three cells with filamentous algae covered coral skeleton and CCA6 and CCA8. No controls. Eighty-one day exposure to substrates.

191 to 197 7-26-98 2-3-99 to 3% (1 of30) settlement. Ten larvae placed in each of 2-9-99 three cells with filamentous algae covered coral skeleton and CCA6 and CCA8. No controls. Ninety-five day exposure to substrates.

Table 2.7. Natural settlement ofM capitata larvae observed over time at settlement survey sites in Kane'ohe Bay. Estimates extrapolated to mean number m-2 ofcolonizable reef over time. Estimated Mean % Two-Year Total of No. No. New Settlers Estimated Mean No. No. New Settlers Plate m-2 coIonizable New Settlers m-2 Census Date Observed Sets substrate Oct 1999 41 33 131 1.19 Feb/Mar 2000 6 84 8 0.07 Jul2000 6374 168 3994 36.23 Oct 2000 118 168 74 0.67 Mar/Apr 2001 9 168 6 0.05 Jul/Aug 2001 10872 168 6812 61.79

33 Discussion

Settlement Preference

The complete absence oflarval settlement in control cells lacking substrates in all but one ofthe laboratory experiments (one larvae settled in the GABA experiments after

an extended period oftime), in comparison to the generally high settlement levels in cells with substrates, strongly suggests that an external cue(s) stimulates settlement and metamorphosis ofMontipora capitata larvae. However, laboratory and field observations

failed to demonstrate M capitata settlement preference for a single macroscopic biological substrate, as has been shown/suggested for various Atlantic and Pacific coral species (including members ofthe family Acroporidae; Morse et al. 1988, 1994, Morse and Morse 1991, Morse et al. 1996;Table 2.8). Montipora capitata larvae did show preference for some hard substrate types over others, with at least two encrusting species ofcrustose coralline algae (CCA6, CCA8) being favored over unconditioned rubble and branching coralline algae (CCA1) in laboratory experiments. In addition, recently broken settlement plate fragments tended to be less favored than select species ofCCA and filamentous algae covered dead coral substrate, despite length ofparent plate conditioning. However, it should be noted that the plate fragments in Figures 2.3 and 2.4 were created one day prior to preference experiments. Given the nonporous structure of the parent plates, the level ofplate fragment surface conditioning varied tremendously within a sample (see Negri et al. 2001 settlement results related to unconditioned tiles).

Proportional settlement oflarvae on plate fragments created six days prior to preference testing was high for fragments not exposed to antibiotics, and did not differ from that on comparable CCA substrates (Figure 2.6). In the field, M capitata larvae settled on a

34 variety ofencrusting community species as well as conditioned, uncolonized tile surfaces. Identification ofsettlement substrate types by itselfcannot verify a specific chemical inducer community, as settlement may occur on alternate substrates after larval contact with inducer organisms (Morse et al. 1988, Morse et al. 1996, Heyward and Negri

1999, Negri et al. 2001). However, the distribution ofM capitata across a wide range of habitats (Maragos 1972, 1977, Kolinski and Jokie12002) and the absence ofobservations ofany specific preference in the laboratory and field suggests that, ifM capitata settlement is chemically induced, the biological community(s) producing the chemical inducer(s) is likely to be diverse and/or fairly ubiquitous in nature.

Marine bacteria have been noted as inducers oflarvae settlement in a variety of marine invertebrate species (reviewed by Johnson et al. 1997, Leitz 1997, and Unabia and

Hadfield 1999, Negri et al. 2001) including at least one species ofscleractinian coral

(Negri et al. 2001, see also Harrigan 1972 and Golbuu 1996). In some cases, associations between larval settlers and specific encrusting macroscopic organisms have been found to depend on the presence ofsurface bacteria (Johnson et al. 1991, 1997, Johnson and

Sutton 1994, see also Negri et al. 2001). Laboratory experiments involving the antibiotic chloramphenical strongly supported the hypothesis ofa role for bacteria in the induction ofM capitata larvae settlement. The experiments suggested the antibiotic directly affected the prokaryotic community and had no direct negative effect on larval ability to settle and metamorphose. Antibiotic effects on settlement likely resulted from a reduction in the entire prokaryotic community including community members that induce M capitata settlement. Alternatively, antibiotics may have acted to change the natural bacterial community structure in a manner that allowed for a subsequent disproportionate

35 increase in bacteria that inhibited settlement (see Johnson et al. 1997). This alternative is

based on the premise that antibiotics never completely eradicate bacterial populations,

which may recover as antibiotic concentrations decline. Nevertheless, differences

associated with settlement levels at particular sites (Kolinski Chapter 3) and orientations

may reflect different bacterial compositions inhabiting substrate surfaces.

Experimentation on the efficacy ofisolated strains ofbacteria to induce settlement should

help to further clarify the likelihood ofthe inducer hypothesis and would be the next

logical step in isolating and identifying potentially active inducer molecules and

associated community surface distributions.

Gamma aminobutyric acid (GABA) does not appear to playa stimulus role in the

settlement ofM capitata larvae despite its production, release and/or uptake by some marine bacteria (Imhoff 1986, Mountfort and Pybus 1992a, 1992b, Kaspar and Mountfort

1995, Johnson et al. 1997). GABA has been shown to induce settlement in larvae of various marine invertebrates (Morse et al. 1979, Heslinga and Hillmann 1981, Morse

1990, Rumrill and Cameron 1983, Pearce and Scheibling 1988, Nadeau et al1989, Avila

et al. 1996); however, other tested cnidarian larvae, including Agaricia spp. (Morse et al.

1988), Alcyonium siderium (Sebens 1983), and Phialidium gregarium (McCauley 1997),

also failed to settle in response to GABA exposure. GABA concentrations at or above

10-3 M were found to be toxic to M capitata larvae. Although controversy exists as to the mechanism by which GABA influences invertebrate larvae to settle (Pawlik 1990, 1992), it has been suggested that inducement results partially through inhibition ofneural

activity related to ciliary movement (Rumrill and Cameron 1983, Barlow 1990, Pawlik

1990, Mountfort and Pybus 1992a, 1992b). Qualitative observations suggested 10-3M and

36 10-4M GABA treatments reduced M capitata larval swimming activity. Ifsuch action was the direct result ofinhibition ofcilia functioning, then cessation ofcilia movement alone is not enough to stimulate settlement and metamorphosis in M capitata larvae.

Table 2.8 summarizes available information related to coral settlement induction and catalogues scleractinian species in terms oftheir reported larval settlement preference behavior as generalists or specialists, highlighting identified or presumed inducer chemicals and/or communities. The relevance ofdistinguishing these groups is at least three fold. First, such categorization may help elucidate individual species distributions and recruitment rates (Morse et al. 1988, Morse 1992, Carlon and Olson 1993, Morse and

Morse 1996, Raimondi and Morse 2000, Heyward and Negri 1999). Second, limited or specialized preference highlights potential life-history vulnerability with regards to physical and/or chemical benthic anthropogenic impacts. Third, limitations in natural recovery ofdegraded reefareas and method options to artificially accelerate reefrecovery may be better understood through clarification ofthe specifics oflarval settlement­ substrate dependence (Negri et al. 2001). Available information to date is extremely limited and includes reported species preferences that require more thorough investigation. Excluding those corals for which additional types ofsubstrate need to be tested (Potential Preference in Table 2.8), an equal number ofcoral species are listed as generalists and specialists (note: the majority ofspecialists show weak settlement preferences, with variability in substrate selection evident). Given the apparent association ofM capitata larval settlement with marine bacteria and the propensity to settle on a wide variety ofconditioned substrate types, M capitata is categorized as a generalist. Such categorization does not disallow association with one particular species

37 Table 2.8. Suggested classification of scleractinian species based on available information ofobserved larval settlement preferences. Generalist = no strong preference among different types oftested substrates. Specialist = significant preference for a limited number ofsubstrate types. *= Additional types ofsubstrate need to be tested. Species Investigator Preference/Inducer Generalists

Acropora Heyward and Negri -Variety ofsubstrates induce settlement in laboratory millepora (1999) experiments including seven species ofcrustose red algae, two species ofbranching coralline algae, bare rubble, CCA Negri et al. 2001 skeleton and Goniastrea skeleton. -Pseudoalteromonas sp. isolated from CCA Hydrolithon onkodes induces partial metamorphosis in high percentage of larvae, but few attach to substrate (perhaps due to use of artificial seawater). Agaricia danai Morse et al. 1988. -Settlement on mixed CCA species on coral rubble, microalgal and bacterial films on coral substrate, and on boiled coral substrate (microalgal and bacterial films present prior to boiling). Inducer(s) unknown.

Faviafragum Carlon and Olson -No preference in field trials- bare coral rubble, filamentous 1993 (but see Lewis algae coated rubble and coralline algae. Inducer(s) unknown. 1974).

Montipora This study. -High levels ofsettlement on at least three species ofcrustose capitata coralline algae and filamentous algae covered coral skeleton. Role ofmarine bacteria evident in laboratory experiments.

Pocillopora Harrigan 1972. -Most hard substrates tested in laboratory experiments. damicornis Suggested inducer a thin living diatomaceous-bacterial film, particularly in association with green filamentous algae.

Specialists - Strong Preference Agaricia Morse et al. 1988, -Specific preference for three tested species ofCCA, humilis Morse and Morse including Hydrolithon boeresenii and Peyssonnelia sp. 1991 (see also Morse Sulfonated lipoglycosaminoglycon (polysaccharide) isolated et al. 1994), Morse et from walls ofCCA Hydrolithon boergensenii induces 100 % al. 1996. Morse and metamorphosis oflarvae. Pacific CCA Hydrolithon reinboldii Morse 1996, also induces settlement. Induction by specific receptor Raimondi and Morse recognition proposed. 2000. - Weak Preference Agaricia Carlon and Olson -Suggestive preference for rubble coated with CCA agaricites 1993. Paragoniolithon typica and Spongites sp. 2. Minor settlement on bare rubble. A. tenuifolia Morse et al. 1988 -Significant preference for mixed CCA species on coral rubble compared to microalgal and bacterial films on coral substrate, and boiled coral substrate (CCA present prior to boiling). No settlement in seawater only treatments. Non­ stringent CCA preference suggested.

(continued)

38 Table 2.8. (continued) Species Investigator Preference/lnducer Specialists - Weak Preference A. tenuifolia Morse and Morse -Sulfonated lipoglycosaminoglycon (polysaccharide) isolated 1991, Morse et al. from walls ofCCA Hydrolithon boergensenii induces high 1994. level ofsettlement on polystyrene container. Goniastrea Golbuu 1996. -Significant preference for CCA Hydrolithon reinboldii with retiformis limited settlement on CCA Peysonelia sp. and rocks covered with microbial reeffilms. No settlement on reefrock devoid ofCCA and reefmicrobial films.

Stylaraea Golbuu 1996. -Significant preference for rocks covered with microbial reef punctata films with limited settlement on CCA Hydrolithon reinboldii, Peysonelia sp. and reefrock devoid ofCCA and reef microbial films. - *Potential Preference Acropora Morse et a1. 1996. -Acropora digitifera and A. nasuta preference for CCA digitifera Hydrolithon rienboldi with no settlement in brown algae A. hyacinthus Lobophora variegata or seawater only treatments. Settlement A·florida only on CCA in single tank experiment ofall listed species A·formosa given choice ofCCA, live corals, inert reefcollected A. gemmifera substrata, fouled panels and inert settlement plate materials. A. nasuta Suggest strict requirement for contact with CCA. Limited A. tenuis alternative substrates tested in first experiment. Identification A. sp.1,2&3 ofCCA settlers in second experiment not reported (larvae identified, but not settlers).

Acropora Negri et al. 2001. -High levels ofsettlement in presence ofCCA Lithophyllum willisae sp (no settlement on sterile tile or Porites skeleton). Pseudoalteromonas sp. bacteria isolated from CCA Hydrolithon onkodes induces metamorphosis in high percentage oflarvae in presence oftile (mostly unattached) and Porites skeleton TPA (see below) has no effect. Limited alternative substrates tested. Cyphastrea sp. Morse et al. 1996. -Preference for CCA Hydrolithon rienboldi with no settlement on dead coral branches, the brown algae Lobophora variegata or in seawater only. Molecule isolated from Hydrolithon rienboldi and Peysonnelia induces metamorphosis. Suggest strict requirement for contact with CCA. Method description lacking and alternative substrates limited. Favia favus Morse et al. 1996. -Limited laboratory tests found preference for CCA Peyssonnelia sp. with no settlement in the brown algae Goniastrea Lobophora variegata or seawater only treatments. Suggested retiformis stringent requirement for CCA inducer found in Hydrolithon rienboldi and Peysonnelia sp. Limited alternative substrates tested. Unknown Stylophora Henning et al. 1996, -Settlement induced by TPA (l2-0-tetradecanoylphorbol-13­ pistillata Fleck 1997. acetate), a tumor promoting phorbol ester suggesting activation ofthe protein kinase C signal transduction pathway. No natural substrates tested.

39 or strain ofbacteria. However, the evidence to date suggests that whatever the nature of

the relationship, substrate association does not appear confined to any recognizably

discrete type or member ofthe macroscopic biotic community.

No significant difference in levels ofsettlement existed between 3.3 month and

15.8 month field conditioned plates or 15.8 month and 28.3 month plates suggesting the

level ofsubstrate conditioning required for M capitata larval settlement is reached

relatively rapidly and can be maintained over time. This finding supports the tentative

conclusion that marine bacteria serve an inductive role in M capitata settlement, as

bacteria are a fundamental part ofearly marine successional substrate communities that

persist in some form on exposed surfaces. These results also support Fitzhardinge's

(1993) contention that M capitata can be among the early scleractinian colonizers of

recently created substrates, although recruits may be difficult to detect without magnification for some time due to relatively slow rates ofgrowth (Fitzhardinge 1993,

Kolinski Chapter 3). Settlement association with an early successional organism has

advantages for an initially slow growing coral by providing footholds on newly created

surfaces. However, the fact that the inducing community may be widespread and

potentially ubiquitous suggests a generalist's strategy for achieving recruitment. Such a

pattern corresponds well with the high larval dispersal potential and tolerance ofadults

for a wide range ofhabitat conditions (Maragos 1972, Kolinski and Jokiel2002). In

Hawai'i, adult M capitata rank relatively high in their ability to compete for space.

These characteristics apparently allow for relative spatial prominence in the species

depauperate Hawaiian coral community. However, in other areas ofthe Pacific where

species numbers are high and competition for space is strong, the generalist settlement

40 strategy may be the most important means by which M capitata is able to maintain a

presence.

Settlement Competency

Laboratory and field observations indicate that the majority oflarvae of

Montipora capitata can and will settle soon after completing embryogenesis ifthey

encounter a suitable settlement substrate. The ability to settle remains high for a period

approaching six weeks. Small numbers ofM capitata larvae may maintain and/or delay the onset ofsettlement competency for periods oftime extending up to and potentially beyond seven months. Other studies have shown variation in settlement rates ofcoral larvae, with most settlement occurring soon after embryogenesis and development

(Harrison and Wallace 1990, Carlon and Olson 1993, Morse et al. 1996, Zaslow and

Benayahu 1996, Wilson and Harrison 1998, Heyward and Negri 1999, Ayre and Hughes

2000, Harii et al. 2001, Miller and Mundy 2003). Estimates on sc1eractinian coral capacity for long-term larval dispersal are few, with previously reported maximum values for a brooding species reaching 120 days for Alveoporajaponica (Harii et al. 2001) and

105 days for the broadcast spawning species Platygyra daedalea (Nozawa and Harrison

2003; see also Tranter et al. 1982, Richmond 1987a, 1988, Wilson and Harrison 1998).

The observed settlement ofan M. capitata larvae 207 - 213 days following parental release provides the longest duration oflarval competency for a sc1eractinian coral. Given that direct long-term assessments can only, as yet, be made under laboratory conditions, the extent to which maximum reported periods are realized under natural conditions remains unknown.

41 Montipora capitata in Hawai'i has a shorter pre-competent phase oflarval

development compared to some broadcast spawning species in Hawai'i and other parts of

the world (Kojis and Quinn 1981, Shlesinger and Loya 1985, Babcock and Heyward

1986, Harrison and Wallace 1990, Morse et al. 1996, Hayashibara et al. 1997, Wilson and

Harrison 1998, Schwarz et al. 1999). Larvae in embryonic and early development stages

presumably lack control over their vertical position within the water column and thus

disperse wherever currents may take them. A short period ofdispersal enhances potential

for some local retention, especially in protected areas influenced mainly by tidal currents.

However, the capacity ofM capitata larvae to maintain competency for up to six weeks

allows for reefconnectivity among both near and distant shorelines within the Hawaiian

island chain (see Grigg 1983). High population densities ofM capitata (in bays, inlets

and along naturally protected shorelines) enhance probabilities offertilization and may be

critical to long-term species persistence in more exposed areas by providing sources of

larvae. Hodgson (1986) identified large concentrations ofM capitata larvae in tow

samples within, and to some extent outside, Kane'ohe Bay and proposed an overall mass

efflux ofcoral larvae from Kane'ohe Bay on the order of90,000 larvae per day within

the coral spawning season.

Planktonic survival and competency oflarvae will be affected by numerous

factors including temperature, food availability and predation (Thorson 1950, Pechenik

1987, 1999, Hoegh-Guldberg and Pearse 1995, Morgan 1995, Zaslow and Benayahu

1996, Pratchett et al. 2001, see also Jackson and Strathmann 1981). Measured daytime

water temperatures in this study ranged between 28.1 to 21.4°C over the seven-month period. Pre-competent and/or competence periods may be less in warmer waters

42 (Pechenik 1987, 1990, Hoegh-Guldberg and Pearse 1995, Zaslow and Benayahu 1996).

Unlike many other coral species, corals in the genus Montipora receive maternally derived zooxanthellae (Richmond 1997). It is likely that photosynthetically fixed carbohydrates help maintain metabolic activities in these larvae over time (see Richmond

1981, 1987a, 1988). Additional mechanisms ofacquiring nutrition may include uptake of dissolved organic matter and direct feeding (Boidron-Metairon 1995). Larval feeding has not been examined in this species, however buoyant polyp balls (Richmond 1985a,

Zaslow and Benayahu 1996) with apparently functioning tentacles have been noted in rearing bottles (Kolinski pers. obs.). Predation probabilities are based on predator abundance, predator preferences, larval abundance, larval behavior, availability of alternative prey, and time of exposure. Virtually nothing is known about natural rates of predation on larvae ofM capitata.

Although only low numbers ofM capitata larvae appear capable ofmaintaining an ability to settle after extended periods oftime, the long-term results ofthis study provide a mechanism for coral connectivity between distant islands and archipelagos.

Maragos and Jokiel (1986) estimated surface water transport times from Johnston Atoll

(800 km southwest ofHawai'i) to Hawai'i to range between 21 and 50 days (well within the competency period ofhigh proportions ofM capitata larvae) and 40 to 50 days from

Hawai'i to Johnston. Montipora capitata is common to low water motion areas at

Johnston Atoll (Jokiel and Tyler 1993). Grigg (1981) estimated a surface water drift time of187 days from Wake Island to Hawai'i (2700 km). Jokiel (1984) noted that a small boat lost offthe island ofHawai'i was recovered 202 days later 7500 km distant in the

Philippines (see Honolulu Advertiser, May 1, 1982) and suggested this direction for long

43 distance reefconnectivity (Jokiel and Martinelli 1992). Larval connections between

Hawai'i and southwest Pacific Island regions are thus reasonable for this species,

especially ifviewed over geologic timescales. Connectivity may also occur via the rafting

ofsettled corals on floating debris (JokieI1984, 1990a). Such means oftransport greatly

extends the potential period and breadth ofdispersal as recruits may grow to reproductive

size while remaining buoyant (JokieI1990a). The generalist settlement behavior ofM capitata makes rafting a feasible mechanism oftransport for this species over geologic time. Unidentified species ofMontipora have been found attached to pumice within the present geographic range ofM. capitata (JokieI1990b).

Long distance larval drift has been used to account for coral species connections and/or founding events across the east Pacific Barrier (Dana 1975, Richmond 1987a,

1990), and may help maintain populations ofvarious broadcast spawning species at certain high latitude locations (Veron 1995, Wilson and Harrison 1997, 1998). Although

M capitata has an extensive larval competency period, displays generalist settlement tendencies, and, following Veron (2000b), is widespread throughout much ofthe Indo­

Pacific, it presently appears to be absent west ofIndonesia, east ofHawai'i, and reaches its eastern south Pacific extreme just past Fiji (Veron 2000b). Maragos (1995) suggests an even more restricted range, limited to the Hawaiian, Johnston and Line Islands. The

Hawaiian-Line Islands region does have a high proportion ofcoral endemics (Maragos

1995, Maragos et al. in press), which suggests isolating mechanisms that may be related to distance from other island regions. Either way, the absence ofM capitata beyond stated boundaries may result from a combination offactors such as species age and origin

(Stehli and Wells 1971), environmental suitability for recruits and adults (Dana 1975),

44 composition ofcompeting benthic communities, current dynamics from likely source

areas, and rarity ofintroduction due to low larval probabilities ofreaching competency

extremes and suitable destinations (see Jackson and Strathmann 1981). The biogeography

and associated systematics ofscleractinian corals are contentious issues not easily resolved. Additional research is needed.

Pechenik (1990) noted variability in periods ofmetamorphic competence is common to laboratory cultures ofmost marine invertebrates and suggested relevance in distinguishing delayed onset ofcompetence from post-competent delays in metamorphosis. Either mechanism implies the capacity for dispersal, but variation in timing suggests slightly different strategies for affecting dispersion scale. Competent larvae that delay metamorphosis possess ability to settle as soon as a suitable substrate is encountered. Variation in settlement is likely substrate driven, and the scale ofdispersal may be best explained by suitable substrate availability. However, variation in the delay ofthe onset ofcompetency implies a potentially inherent forcing mechanism for wide scale dispersion (see Harper 1977, Raimondi and Keough 1990). Such variation is likely to be genetically/developmentally driven, and may not negate the possibility for further post-competent, substrate dependent, metamorphic delay. Wilson and Harrison (1998) observed variation in delay ofsettlement and metamorphosis in the presence ofsubstrate for larvae ofthree scleractinian species ranging 26 to 78 days after first recorded settlement. In all above experiments involving M. capitata larvae, variation in timing of settlement was evident and at least a few larvae remained motile. The longest-term larvae were exposed to substrates 32 to 95 days prior to their settling. Early onset ofcompetency does appear to be proportionately high in M capitata larvae, and variability in settlement

45 rates may be due to variation in the degree or type oflarval cue specificity (Raimondi and

Keough 1990), or a host ofother factors. However, anecdotal evidence suggesting potential pre-competent period variation cannot be ignored and should be fully investigated. Given the prolific reproductive nature ofboth broadcast spawning and p1anu1ating coral species, variation in periods ofpre-competence may be an ecologically advantageous and relatively inexpensive means by which to enhance wide-scale offspring dispersion.

46 CHAPTER 3

SETTLEMENT, SURVIVAL AND EARLY GROWTH RELATIVE TO OTHER

COMMON HAWAIIAN SCLERACTINIANS

Introduction

Coral reefs are dynamic ecosystems that are subject to disturbances ofboth natural and anthropogenic origin at a variety ofspatial and temporal scales (Connell

1978, 1997, Brown and Howard 1985, Grigg and Dollar 1990, Maragos 1993, Jokiel et al. 1993, Richmond 1993, Brown 1997, Hughes and Connell 1999, Sousa 2001).

Disturbance provides potential to disrupt an existing balance ofcompetitors, in many cases creating habitats or conditions suitable to the arrival and survival ofnew settlers

(Grigg and Maragos 1974, Connell 1978, 1997, Pearson 1981, Grigg 1983, Connell and

Keough 1985, Karlson 1999, Sousa 1984,2001). Reefcorals in general reproduce and disperse via planktonic larvae that are capable ofestablishing individuals in areas with suitable substrate and conditions. The (re)establishment, long-term maintenance, and/or continued dynamics ofmany ofthe species members of any coral community will be dependent to a large extent on larval influxes and the early life history success of resultant settlers (Pearson 1981, Wallace 1985, Harrison and Wallace 1990, Richmond

1997, Hughes and Tanner 2000, Underwood and Keough 2001).

Natural settlement and recruitment ofvarious coral genera has been monitored intermittently throughout the Pacific (Harrison and Wallace 1990 and cited references,

Fisk and Harriott 1990, Sammarco 1991, Harriott 1992, 1999, Glynn et al. 1994, 1996,

2000, Gleason 1996, Harriott and Banks 1995, Banks and Harriott 1996, Baird and

Hughes 1997, Holmes et al. 1997, Connell et al. 1997, Dustan and Johnson 1998, Fairfull

47 and Harriott 1999, Hughes et al. 1999,2000,2002, Kojis and Quinn 2001). However, characteristics ofsettlement, survival and early growth are not well documented for most

Indo-Pacific sc1eractinian coral species. Previous efforts in Hawaiian waters have provided insights on early life history characteristics ofa few ofthe more common corals

(Dollar 1975, Polacheck 1978, Fitzhardinge and Bailey-Brock 1989, Fitzhardinge 1993,

Deemers 1996). Such insights, however, have been limited and/or remain somewhat speculative due to either the short-term nature ofthe studies, an absence offocus on

"invisible recruits" (sensu Wallace 1983), and/or an inability to track the fate of individuals across observation periods. Overall, there is a general deficiency of information on coral recruitment dynamics in Hawaiian waters (Kolinski and Cox 2003).

Montipora capitata is a coral common to a variety ofHawaiian reefhabitats

(Maragos 1972, 1977, Kolinski and Jokie12002, Maragos et al. in press). This species is a highly synchronized summertime spawner with buoyant egg-sperm bundles, surface fertilization (Heyward 1986, Kolinski and Cox 2003), a two to three day embryonic period (Mate et al. 1998, pers. obs.), and a propensity for settlement soon after embryogenesis is complete (Kolinski Chapter 2). Similar to a number ofother coral species, M. capitata is a generalist in terms ofits preference for settlement on various hard substrates (Kolinski Chapter 2). Montipora capitata is the only coral species within

Kane'ohe Bay that gives rise to large, very obvious, summertime surface slicks of gametes and developing embryos on the nights and mornings following synchronized spawning. Plankton tows within the bay found M capitata larvae to be at least an order ofmagnitude more concentrated than larvae ofany other coral species following respective species spawning events (Hodgson 1986). Early studies, however, have

48 suggested relatively low levels ofsettlement and/or survival at monitored sites within the bay (Polacheck 1978, Fitzhardinge and Bailey-Brock 1989, Fitzhardinge 1993).

This study focused on settlement and early life history characteristics ofM. capitata in the shallow waters ofKane'ohe Bay in an effort to elucidate the enigma of high levels oflarval production and low levels ofrecruitment. Settled individuals were tracked and measured on artificial plates over a two-year period. Specific comparisons of early life history characteristics ofM. capitata were made relative to other common, naturally settling Hawaiian scleractinian species as a means to understand potential similarities and differences in early life history strategies.

Methods

The term "settler" was used in this study to refer to an individual coral that settled, as a larva, on a settlement plate apparatus. This terminology encompassed metamorphosed individuals and was used for settled corals throughout the duration ofthe study.

Settlement Plates

Field observations ofscleractinian coral settlement were made on settlement plate sets established at six shallow water (depths 1.5 to 3.5 meters) sites, including one fringing (F) and patch (P) reef each in north (N), central (C) and south (S) Kane'ohe Bay

(Kolinski Chapter 2, Figure 2.1). Plate sets are described and diagramed in Chapter 2

(Figure 2.2). Initial plate sets (two to a post, seven posts at a site) were installed in

March, three months prior to the 1999 Montipora capitata summer spawning, with replicate sets being added three months prior to the 2000 summer spawning. Plate sets removed for analysis were always kept submerged in individual containers and were

49 transferred by boat to the Hawai'i Institute ofMarine Biology (HIMB) for settler

identification, measurement and mapping using a Wild Heerbrugg dissecting microscope

(with the aid ofa 10 x 10 cm monofilament sectioned quadrat). In all cases, halfofthe

sets on each post at a site were collected at one time and analyzed, with collection

schedules rotated among sites to ensure equivalent time-related proportionate sampling.

Plates were held in individual containers in a shaded flow-through seawater table while in the laboratory. Container seawater was filtered (40 11m) and changed regularly. These procedures ensured that plates were not contaminated with larvae entering with seawater

at the lab. All plate sets were returned to their respective posts within one to two days.

Examination ofall plate sets at each census was conducted in less than one month.

In June 1999, roughly halfofthe plate sets (one from each offive to six posts at a site) were inspected and then seeded with cultured M capitata larvae (Kolinski Chapter

2) for related experiments. These sets were enclosed within individual Glad™ brand plastic transparent disposable containers (739 ml) modified with 40 11m Nitex screen on all but the top surface to allow for adequate water exchange (Kolinski Chapter 2, Figure

2.2). The PVC clips ensured that chambers (including screens) did not touch the plates.

Five hundred 6-day-old larvae were injected into each chamber in the field. Chambered sets were collected and settlers mapped and scored after 6 days. These seeded plates were returned (without the chambers) to the field and were inspected with the other plates the following fall, winter-spring and summer through summer 2001.

Settlement Among Sites

Adult colonies ofvarious coral species were placed in separate flow-through seawater tables at HIMB for observation ofspawning times (see Kolinski and Cox 2003)

50 and collection ofgametes for larval culturing and settlement. Larvae ofknown species

were allowed to settle on seawater conditioned (three months) unglazed terracotta tiles,

and these laboratory settlers acted as references in the species identification offield

settlers. Notes on species characteristics, digital photos and measurements were taken of

laboratory settlers over time. Successful settlement and survival occurred for larvae of

Montipora capitata, M patula, Porites compressa, P. lobata, Fungia scutaria,

Pocillopora damicornis, Cyphastrea ocellina, and Tubastraea coccinea. Photos of

settlers provided in Fitzhardinge (1993) and English et al. (1997), and live polyps ofadult

colony fragments, were also referenced to aid in species identifications.

Cumulative counts ofall field settlers for each species at each site were made for

general comparison. A factorial ANOVA (with Tukey HSD means comparisons, a =

0.05), involving only species with observed settlers at all sites, was used to examine

differences between mean numbers ofsettlers m-2 (lOglO transformed) observed over two

one-year monitoring periods. The comparison was limited to two-way interactions due to

a restrictive number ofdegrees offreedom.

Settlement Over Time

Seasonal settlement was examined for those species having enough settlers at multiple sites to allow gross patterns in settlement timing to be discerned. Mean numbers ofsettlers m-2 (+1 10glO transformed) ofeach species in each season ofobservation

(summer = July to early August, fall = October, winter-spring = February to April) over a two-year monitoring period were determined for graphical comparison. Large site variability between species within and across seasons prohibited direct inclusion ofall

species in a single factorial ANOVA. Thus, each species was analyzed using separate

51 mixed factorial ANOVAs (with Tukey HSD means comparisons, a = 0.05), analyzing the

2 mean number ofsettlers m- • The results were considered in relation to the reproductive mode and observed spawning patterns ofeach species.

Settler Survival

Montipora capitata

Proportional survival ofone to two-week old seeded (1999) and natural (2000)

summer Montipora capitata settlers was plotted against time to assess general trends in

survivorship (Deevey 1947). Three survivorship curves were constructed; one for seeded larvae (1999), and one each for larvae that naturally settled on three month and 15.8 month (2000) field conditioned plates. Individual comparisons ofproportional survival at three, nine and 12 months post-settlement were made between the groups. Initially, seasonal survival ofnatural summer 2000 settlers on three and 15.8 month conditioned plates was compared using paired t-tests on arcsine square-root transformed proportional survivorship values pooled for each site at three, nine and 12 months (only sites with> 20 individuals per category were analyzed). The findings allowed for all natural 2000 summer settlers at each site to be pooled for similar comparative analyses with seeded

1999 survivorship estimates. Survival was calculated as the proportion ofinitial summer settlers remaining alive at each census. Settlers on clips and wing-nuts were not included in any ofthe survival analyses as these areas were used for plate handling.

Comparison with Other Species

Comparison ofsurvivorship ofMontipora capitata to that ofPorites compressa and Pocillopora damicornis was difficult to express due to protracted settlement times across seasons for the latter two species. Individuals entered into observation

52 progressively over the course ofthe two-year monitoring period and thus were monitored for varying periods oftimes. Unfortunately, variation in timing between censuses resulted in the data not being amenable for curve construction and analysis in available survival analysis statistical packages. A survival analysis framework was, thus, constructed in

Microsoft Excel, accumulating all that was known concerning each individual's fate over time. Time zero was set for each individual at the time it was first observed. Cumulative counts ofeach species at each site were calculated for each of21 months beginning at month 4, and included total number ofindividuals for which information existed, total known number ofsurvivors, and total number ofright censored individuals (i.e., individuals surviving beyond the end ofthe monitoring period, Lee 1992). The first three months were not included in the analysis due to a lack ofinformation concerning mortality and, thus, a bias towards high survivorship for some species. Months 22 to 25 were also not included, as proportions appeared to become biased by a lack ofknowledge concerning fates ofright censored individuals. Only sites with> 20 individuals were included for the construction ofeach species mean survivorship curve. Seeded M capitata settlers were pooled with natural settlers for analysis.

Statistical comparisons ofproportionate survival were made separately at four, 12 and 21 months using one-way ANOVAs and Tukey HDS tests (a = 0.05) on square-root arcsine square-root transformed data. A single outlier was omitted from the 21-month analysis for P. compressa to meet model assumptions (note: test results were similar with and without the outlier). An additional analysis comparing proportional survival offour­ month survivors at 12 and 21 months was made to examine whether potential differences between species in numbers ofvery recent settlers at times offirst observation could be

53 biasing long-term survival comparisons. The data were arcsine square-root transformed and analyzed using separate one-way ANOYA and Tukey HDS (a = 0.05) tests.

Settler Growth

Measurements ofindividual settlers at each census were made using a Wild

Herbrugg dissecting microscope with a lens micrometer and included counts ofpolyps

(when possible), longest measured length, perpendicular width and, for Pocillopora damicornis when appropriate, height. Projected area ofeach settler at each census was calculated as nab (area ofan ellipse; a = Y:z longest measured length; b = Y:z perpendicular width). Surface areas were not estimated for growth comparisons due to differences in growth forms and patterns as settlers grew larger (consistent 2 dimensional increases for

Montipora capitata and Porites compressa versus initial 2 followed by 3 dimensional branching growth for P. damicornis). Settlements on clips and wing-nuts were not included in any ofthe growth analyses as these areas were used for plate handling.

Growth was estimated for select individuals as the change in polyp numbers over time. Only individuals with zero and/or positive increases in numbers ofpolyps in three or more consecutive counts were selected for analysis, with one set ofzero and/or positive consecutive measurements and corresponding times being included for each individual. Polyp counts were standardized by dividing each count by the initial number ofpolyps in each set (as each polyp in a small colony had an ability to divide, this conversion standardized growth to an individual polyp for comparison). All standardized counts were logro transformed. Slopes were estimated using regression analysis (Statistix

1.0). Growth estimates from all sites were pooled for interspecific comparisons using a one-way ANOYA. Consistency in growth was assessed between sites having replicate

54 M. capitata, P. compressa and P. damicornis slope estimates (Central Fringing, Central

Patch and North Patch reefs) using a mixed factorial ANOVA. Post-hoc comparisons

were made using Tukey HDS tests with a = 0.05.

The relationship between settler growth and survival was examined for M

capitata, P. compressa and P. damicornis by comparing slope estimates ofgrowth as

detennined above (pooled across all sites) between right-censored "survivors" (i.e.,

individuals that survived beyond the end ofthe monitoring period) and similar aged (nine months plus) settlers known to have died over the course ofthe monitoring period. A

factorial ANOVA, with growth as the dependent variable, and a binomiallogit homogeneity ofslopes model, with survival as the dependent variable, were used to

assess the relationship. Within species contrasts were made using Bonferroni corrected

P-values.

Mean and minimum time estimates for M capitata, P. compressa and P. damicornis settlers to reach 1 cm2 projected area in the field were made using the mean and maximum growth slope estimates for right-censored individuals ofeach species.

Regression analysis was used to predict the mean number ofpolyps for a 1 cm2 projected area for each species. The polyp values were 10glO transfonned and placed in respective mean growth regression models, which were then solved for time.

Habitat Characteristics

Water Quality

Water motion, total suspended solids, sedimentation rates, salinity and temperature were measured at center and end settlement plate posts (n = 3) at each site from July 1999 through October 2000. With the exception ofsalinity and temperature,

55 times between consecutive measurements always exceeded one month, with dates

selected in advance on a haphazard basis. Water motion was determined using Plaster of

Paris "clod cards" (Doty 1971, Jokiel and Morrissey 1993), with two clod cards being

anchored to each ofthree bricks at each site per trial for a total ofnine 24-hour trials.

Control cards were placed in a five-gallon (18.92 l) bucket ofseawater in a shaded, wind

protected area on land. The cards were briefly rinsed in still fresh water, air-dried in an

air-conditioned, dehumidified building for a month and then weighed on a Mettler P160

balance. Three one-liter water samples were collected at each site a total ofeight times

for measurement oftotal suspended solids (TSS). The water was collected adjacent to

each ofthree posts in clear one-liter Nalgene polyethylene bottles. Sample volumes were

vacuum filtered onto pre-weighed, 500°C pre-ashed GFF filters, with ample rinsing using

de-ionized water. The TSS samples were placed in a 60°C drying oven for 24 hours prior to being weighed on a Mettler AB 104 electronic microbalance. Rates ofsedimentation were measured over 10 separate 24-hour periods using 15.24 cm deep, 5.45 cm wide, bottom-capped polyvinyl chloride (PVC) pipe traps. One trap was clip-tied to each of three posts at each site with openings level to plate top surfaces. The traps were plugged prior to removal from the posts and were vacuum filtered onto individual pre-weighed,

500°C pre-ashed GFC (8 months) and Whatman filters (2 months), with ample rinsing using de-ionized water. The sediments were dried at 60°C for 24 hours and then weighed on a Mettler ABI04 electronic microbalance. Samples ofwater adjacent to three posts per

site were collected in 125 ml clear Nalgene bottles and rapidly measured for salinity and temperature in the field using either a YSI Model 30 conductivity meter or refractometer and thermometer. Salinity was measured on 15, and temperature on 9 occasions.

56 Values ofall measured water quality parameters were averaged across trials for

inclusion in a Pearson's correlation matrix for exploratory analysis against measurements

ofMontipora capitata, Porites compressa and Pocillopora damicornis mean yearly

settlement, growth and 12-month survival using the following transformations: number of

settlers m-2 and water motion 10glO transformed; growth = slopes ofloglO standardized polyp increase per month; percent survival and percent adult cover m-2 arcsine square­ root transformed; total suspended solids and sediment +Il0glO transformed.

Benthic Cover

Benthic composition was assessed along four 10m transects placed within a 10m x 50 m settlement plate zone (nearest transects roughly 15 m distant). One square meter quadrats delineated into 10 dm2 grid cells were used to make 10 contiguous measurements along the length ofeach transect. The measurements were visual estimates ofpercent cover within each quadrat ofsand, rubble, un-colonized hard substrate, coral and algae by species, sponges, and other space consuming macro-invertebrates. The mean percent cover ofMontipora capitata, Porites compressa and Pocillopora damicornis

(arcsine square-root transformed) at each site was included in a Pearson's correlation matrix for exploratory analysis against measurements ofthese species mean yearly

2 settlement m- (lOglO transformed), mean rates ofgrowth (slopes ofloglO standardized polyp increase per month) and 12-month survival (arcsine square-root transformed).

Site Projection Example

Mean values ofsettlement, survival and colonizable substrate were used to project and compare large-scale levels ofsettlement and the four, 12 and 21 month fate of summer 2000 Montipora capitata, Porites compressa and Pocillopora damicornis settlers

57 throughout each 500 m2 study area. Numbers ofsurvivors at 12 and 21 months were

based on estimates ofproportionate survival offour-month survivors so as to limit

potentia110ng-term bias arising from differences in proportions ofvery recent settlers while accounting for right-censored individuals. These species have been observed to

settle on rubble (Kolinski pers. obs.); thus, substrate suitable for colonization was defined

as hard and rubble structure not covered by sand, live coral, macroalgae, sponges,

colonial tunicates, or other space limiting macro-invertebrates.

Results

Settlement Among Sites

Nine identifiable species ofsc1eractinian corals were observed on settlement plates over the two-year period (Table 3.1). Montipora capitata cumulatively accounted for 91 % ofall noted plate settlers and dominated (82 to 96 %) at all but the North Bay sites (roughly 2 % at each site). Porites compressa dominated cumulative counts at North

Patch reef(60 % ofnoted settlers), and Pocillopora damicornis at North Fringing reef

(61 %). Remaining identified species accounted for less than one percent ofnoted cumulative site settlers with the exception ofCulicia tenelfa at North Fringing reef(4 % ofnoted settlers) and Cyphastrea ocellina at North Patch reef(3 %). Montipora capitata,

P. compressa and P. damicornis were the only species observed on plates at all monitored sites. Cuficia tenelfa was observed on plates at all sites with the exception of

South Patch reef(Table 3.1).

Statistical comparison ofnumber ofobserved settlers ofthe three main species,

M capitata, P. compressa and P. damicornis, over two consecutive one-year monitoring periods identified significantly higher levels ofoverall settlement at Central Bay sites,

58 Table 3.1. Cumulative numbers ofscleractinian corals observed on settlement plates at six sites in Kane'ohe Bay from summer 1999 to summer 2001 (C = Central; N = North; S = South; F = Fringing reef; P = Patch reef). Cumulative Number of Settlers Observed % of all noted Species NP NF CP CF SP SF Total settlers Montipora capitata 7 6 1982 14974 260 421 17650 90.87 Porites compressa 276 108 92 352 46 59 933 4.80 Pocillopora damicornis 140 224 137 171 6 10 688 3.54 Culicia tenella 3 15 17 25 1 61 0.31 Cyphastrea ocellina 13 3 16 0.08 Montipora patula 2 2 0.01 Fungia scutaria 1 0.01 Leptastrea bottae 1 0.01 Tubastraea coccinea 1 1 0.01 Unknown 23 15 10 15 4 4 71 0.37 Total 463 369 2242 15539 316 495 19424 % ofTotal 2.38 1.90 11.54 80.00 1.63 2.55

which was mainly the result ofhigh numbers ofM capitata settlers. Although, overall

mean numbers ofM capitata and P. compressa settlers were significantly greater than P.

damicornis, this varied among individual sites (Table 3.2, Figure 3.1). Only P. compressa

displayed statistically similar levels ofcumulative settlement across all sites over the

course ofthe two-year monitoring period (Table 3.2).

Table 3.2. Factorial ANOVA (with Tukey HSD means comparisons, Ci. = 0.05) ofM capitata, P. compressa and P. damicornis settlers (lOglO mean number settlers m-2 coIonizable reef) observed over two one-year monitoring periods (C = Central; N = North; S = South; F = Fringing reef; P = Patch reef; Mcap = M captitata; Pcom = P. compressa; Pdam = P. dam icornis). Df SS MS F P Comparison Intercept 1 198.1377 198.1377 3508.940 0.0000 Year 1 0.1622 0.1622 2.872 0.1210 Site 5 9.5887 1.9177 33.962 0.0000 CF > CP > SF NP NF SP Species 2 1.9663 0.9832 17.412 0.0006 Mcap Pcom > Pdam Year*Site 5 1.7095 0.3419 6.055 0.0078 No within site differences. Year*Species 2 0.2096 0.1048 1.856 0.2064 Site*Species 10 14.0030 1.4003 24.799 0.0000 See below. Error 10 0.5647 0.0565

Site*Species: NF Pdam Pcom > Mcap NP Pcom Pdam > Mcap CF Mcap > Pcom Pdam

CP Mcap > Pdam Pcom SF Mcap Pcom Pdam SP Mcap Pcom > Pdam

Mcap CF CP SF SP NP NF Pcom CF NP NF CP SF SP Pdam CF NF CP NP SF SP

59 5 4.5 • M. capitata ~ P. compressa ('\J---- 4 E U> 3.5 D P. damicornis '- a> ::::a> 3 CJ) 02.5 z ~ 2 a> ~ 0 1.5 ...... 8' 1 .....J 0.5 o NF NP CF CP SF SP Site 2 Figure 3.1. LoglO mean number M capitata, P. compressa and P. damicornis settlers m- colonizable reef(± S.E.) observed over two one-year monitoring periods at sites in Kane'ohe Bay (C = Central; N = North; S = South; F = Fringing reef; P = Patch reef).

Settlement Over Time

Seasonal patterns ofsettlement were discernable for Montipora capitata, Porites

compressa, Pocillopora damicornis and Culicia tenella at multiple sites within Kane'ohe

Bay where overall settlement levels were relatively high (Figure 3.2). Although direct

statistical comparisons among species could not be made due to tremendous site variability within and across seasons, statistical evaluation oftemporal trends for each

species allowed for an indirect comparison (Table 3.3). Three patterns ofsettlement were

apparent. Pulsed settlement was evident soon after peak summer spawning periods in

June and July for the broadcast spawning hermaphrodite M capitata, with the vast majority ofsettlers (> 70 %) appearing to have recently settled. Observed settlement was

significantly higher in the summer, followed by fall and then winter/spring (Table 3.3).

60 Summer settlement ofP. compressa, a gonochoric broadcast spawning species with variable reproductive release from June through September, did not differ significantly from monitored settlement in fall, although settlement in summer and fall was significantly higher than that noted in winter/spring (Table 3.3). Pocillopora damicornis, a brooding species with year round reproduction, did not display statistically significant seasonal differences in settlement. Observed settlement for C. tene/fa did not differ statistically between seasons. The reproductive mode of C. tene/fa is presently unknown.

There was no evidence oflarge discrete settlement pulses within a season for P. compressa, P. damicornis or C. tene/fa during the two-year monitoring period.

4.5 M. capitata ...... -+ P. compressa 4 2ro ~ P. damicornis U) .0 ~3.5 - +- C. tenel/a Q) :0 ro 'cN 3 0 <5 0 N 2.5 E III --~ ...... ~. Q) ~ 2 rJ) ci z 1.5 c ro Q) ~ "U 1 Q) iii E ~ 0.5 UJ -...... o ~ O-t--~--...------L._-.------;------.-----,------, ....J Fal-99 W/S-OO Sum-DO Fal-OO W/S-01 Sum-01 Season Figure 3.2. Temporal settlement (loglO mean number ofsettlers m-2 colonizable substrate +1, ± S.E.) ofselect species ofscleractinian corals at sites in Kane'ohe Bay (M capitata at CF, CP, SF and SP; P. compressa at all bay sites; P. damicornis at NF, NP, CF and CP; and C. tene/fa at NF, CF and CPo Fal = Fall, W/S = Winter-Spring, and Sum = Summer).

61 Table 3.3. Mixed model ANOVAs (with Tukey HSD mean comparisons, a = 0.05) examining temporal differences in mean estimated settlement m-2 ofcolonizable substrate at select sites in Kane'ohe Bay for individual scleractinian coral species (C = Central; N = North; S = South; F = Fringing reef; P = Patch reef). Montipora capitata Source (FIR) df SS MS F P Season Fixed 2 26.3740 13.1870 47.5189 0.0002 Site Random 3 6.4749 2.1583 7.7774 0.0172 Season*Site Random 6 1.6651 0.2775 2.1875 0.1171 Error Fixed 12 1.5223 0.1269 Summer > Fall > Winter-Spring; CF CP>SF SP Porites compressa Source (FIR) df SS MS F P Season Fixed 2 3.6938 1.8469 24.3045 0.0001 Site Random 5 3.4003 0.6801 8.9493 0.0019 Season*Site Random 10 0.7599 0.0760 0.4695 0.8886 Error Fixed 18 2.9133 0.1619

Fall Summer> Winter-Spring; CF NP NF CP SF SP Pocillopora damicornis Source (FIR) df SS MS F P Season Fixed 2 0.8623 0.4312 3.6108 0.0935 Site Random 3 0.3929 0.1310 1.0967 0.4202 Season*Site Random 6 0.7164 0.1194 0.5067 0.7922 Error Fixed 12 2.8276 0.2356 Fall Summer Winter-Spring; =C",-F----",C",-P---"-,N,,,-F----=..;NP"'-. Culicia tenella Source (FIR) df SS MS F P Season Fixed 2 0.5909 0.2955 1.2453 0.3798 Site Random 2 0.4601 0.2301 0.9696 0.4536 Season*Site Random 4 0.9490 0.2373 0.6247 0.6567 Error Fixed 9 3.4185 0.3798 Summer Fall Winter-Spring; CF CP NF

Settler Survival

Montipora capitata

Summer settlers ofMontipora capitata in Kane'ohe Bay cumulatively displayed type III survivorship (Deevey 1947) as shown in Figure 3.3. The number ofmonitored surviving summer settlers declined roughly an order ofmagnitude from initial settlement levels in an approximately three month period for both seeded and naturally settled larvae, and declined an additional order ofmagnitude over the remainder ofa year. Mean proportional survivorship ofnatural (summer 2000) settlers on new and old plates did not

62 10000 - Seeded Plate 99 - 01

New Plate 00 - 01

1000 Old Plate 00 - 01 C) c "S; "~ ::J (f) 100 C/) I- Q) .0 E ::J Z 10

o 3 6 9 12 15 18 21 24 Age (months) Figure 3.3. Survivorship ofnewly settled M capitata larvae over time on three month conditioned plates seeded with larvae (1999) and three (new) and 15.8 (old) month conditioned plates with natural settlers (2000). Data combined for all Kiine'ohe Bay sites. differ significantly at any monitoring period over the course ofa year (Paired t-test on arcsine square-root transformed proportional survivorship at Central Fringing, Central

Patch and South Fringing reefs: fall 2000, n = 3, T = 0.37, P = 0.749; spring 2001, n = 3,

T = 1.03, P = 0.413; summer 2001, n = 3, T = 1.43, P = 0.288). Although mean proportional survivorship ofthe natural summer 2000 settlers was significantly greater than that ofthe 1999 seeded settlers at three months post-settlement (Paired t-test on arcsine square-root transformed proportional survivorship at Central Fringing, Central

Patch, South Fringing and South Patch reefs: n = 4, T = 3.25, P = 0.048), no significant difference was evident at the nine (n = 4, T = 1.33, P = 0.276) or 12 month (n = 4, T =

1.90, P = 0.153) monitoring periods between the two years. Overall survival ofnatural settlers from summer 2000 averaged 2.56 % (± 1.13 S.E., n = 4) by summer 2001.

63 Survival ofsummer 1999 seeded settlers averaged 0.15 % (± 0.08 S.E., n = 6) by summer

2000. The number ofseeded settlers continued to decline to one survivor two years post settlement, with mean survivorship equal to 0.03 % (± 0.03 S.E.; note that an additional seeded settler on a plate wing-nut also survived the duration ofthe monitoring period, but wing-nut settlers were excluded from analysis due to unavoidable mortality caused by removal ofnuts for plate examination). One of 134 (0.74 %) settlers first noted in fall

1999 also survived over the course ofthe monitoring period. None ofsix (0 %) 2000 winter/spring settlers survived to summer 2001.

Comparison with Other Species

Survivorship curves for Porites compressa and Pocillopora damicornis relative to

Montipora capitata settlers from time offirst observation are shown for months four through 21 (limits for determination) in Figure 3.4. Similar to M capitata, P. compressa and P. damicornis displayed Type III survivorship. However, their mean short and long­ term survivorships were significantly greater than that ofM capitata (Table 3.4). A species comparison ofproportional survival ofsurviving four-month old settlers at 12 and

21 months was made to examine whether potential differences in numbers ofvery recent settlers at times offirst observation could be biasing long-term survival comparisons.

Mean proportionate survival offour-month old P. compressa settlers was significantly higher than M capitata at 12 and 21 months (one-way ANOVA ofarcsine square-root transformed data: 12 months, n = 16, F2, 15 = 4.86, P = 0.027; 21 months, n = 15, F2, 14 =

8.39, P =0.005), but neither P. compressa nor M. capitata differed from P. damicornis at either time interval.

64 M. capitata 0.6 P. compressa 0.5 I P. damicornis 0. 0.4 E ~ o .~ 0.3 ~ :::J 1.... (f) ··X... 0.2 .hX. f .... j ..... 0.1

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Months Figure 3.4. Mean proportional survivorship (± S.E.) ofM capitata (all sites), P. compressa (all sites) and P. damicornis (Central and North Bay Fringing and Patch reef sites) settlers observed in Kane'ohe Bay.

Table 3.4. Species comparison ofproportionate survival at 4, 12 and 21 months (one-way ANOVA ofsquare-root arcsine square-root transformed data with Tukey HSD mean comparisons, a = 0.05; one outlier eliminated from P. compressa data set at month 21). Time Source df SS MS FP Comparison 4 months Species 2 1.0257 0.5129 51.49 0.0000 Within 13 0.1295 0.0100 Pcom = Pdam > Mcap Total 15 1.1552

12 months Species 2 0.5428 0.2714 25.91 0.0000 Within 13 0.1362 0.Ql05 Pcom = Pdam > Mcap Total 15 0.6790

21 months Species 2 0.5996 0.2998 28.77 0.0000 Within 12 0.1251 0.0104 Pcom = Pdam > Mcap Total 14 0.7247

Settler Growth

Species Comparisons

Mean linearized rates ofstandardized growth ofMontipora capitata, Porites compressa and Pocillopora damicornis settlers pooled across all sites are shown in

65 Figure 3.5. Montipora capitata grew exceedingly slowly and displayed a significantly

lower mean rate ofpolyp increase over time compared to P. compressa and P.

damicornis, which did not differ. Rates ofgrowth did not differ between sites where

comparisons could be made (Table 3.5). The greatest number ofpolyps achieved for an

individual settler within the two-year monitoring period was estimated to exceed 1000 for both P. compressa and P. damicornis but never reached higher than 18 for M capitata

(8.89 months). Montipora capitata settled and reared in the lab reached a maximum 53 polyps over a 41-month (3.4 year) period. However, mean field growth rates were roughly 2.5 times that oflab-reared settlers (Field = 0.0248 ± 0.0033 S.E.; Lab = 0.0101

± 0.0017 S.E.; Two sample t-test ofno difference, df= 127.4, T = 3.96, P = 0.0002).

0.1

1;j 0.09 "2 m 0.08 -0e- m £; 0.07 +-' c (J) 0 ...... o E 0.06 0)'- .3 ~ 0.05 ...... Q) o (J) 0.04 Q) m Q.Q) o t> 0.03 ...... (J) -C 2 g; 0.02 m-o 0::: a.. 0.01 .c ~ e o CD M. capitata P. compressa P. damicornis Species Figure 3.5. Mean growth rates (slopes ofloglO standardized polyp increase per month) ± S.E. ofM capitata, P. compressa and P. damicornis settlers pooled across sites.

66 Table 3.5. Species comparisons ofgrowth rates (slope ofloglO standardized polyp increase by month). One-way ANOVA used for pooled data all sites. Mixed factorial ANOVA using data from Central Fringing, Patch and North Patch reefs, where replicate estimates occurred for M capitata, P. compressa and P. damicornis. Sites Source (FIR) df SS MS FP Comparison All Intercept 1 1.0707 1.0707 528.0246 0.0000 Species 2 0.2586 0.1293 63.7748 0.0000 Pdam Pcom» Mcap Error 262 0.5313 0.0020

CF, CP, NP Intercept Fixed 1 0.4427 0.4427 584.3048 0.0000 Species Fixed 2 0.1122 0.0561 50.0877 0.0000 Pdam Pcom» Mcap Site Random 2 0.0013 0.0006 0.5836 0.5839 Species*Site Random 4 0.0038 0.0009 0.4719 0.7563 Error Fixed 176 0.3533 0.0020

Projected mean time for settlers to reach 1 cm2 (projected area) in the field was roughly 59 months (4.9 years) for M capitata, 20 months (1.7 years) forP. compressa

and 21 months (1.7 years) forP. damicornis. Minimum projected time to reach 1 cm2 based on fastest recorded growth rate was 13 months for M. capitata and approximately

10 months for both P. compressa and P. damicornis. One-cm2 Porites compressa and P. damicornis were observed at 12 month censuses. Montipora capitata was not observed to reach 1 cm2 throughout the two-year monitoring period.

Extended Single Polyp Status

Large numbers ofMontipora capitata field survivors never exceeded the one polyp stage throughout the monitoring period. The percentage ofsurviving plate settlers in a single polyp stage 12 months after first being observed totaled 61 % for M. capitata,

8 % for Porites compressa and 0 % for Pocillopora damicornis. Competitive interactions within respective one-year periods with other encrusting organisms were noted for 70 % ofsingle polyp M capitata, and 70 % ofsingle polyp P. compressa survivors. A decline in polyp number was significantly related to prior competitive interactions for P. compressa (Chi-square = 9.12, df=l, P = 0.003), but not M capitata (Chi-square = 0.05, df=l, P = 0.821; Table 3.6). One-year single polyp status was significantly related to

67 declines in polyp numbers in P. compressa (Chi-square = 8.55, df= 1, P = 0.004), but not

M capitata (Chi-square = 1.01, df= 1, P = 0.315).

Table 3.6. Survivors noted as single polyps 12 months after first observation. M. capitata P. compressaP. damicornis Category n/tot. % n/tot. % n/tot. % I-polyp stage at 12 months 23/38 60.5 10/121 8.3 0/41 0.0 I-polyp survivors which experienced competitive interactions 16/23 69.6 7/10 70.0 % 0.0 I-polyp survivors which experienced previous polyp declines 1/23 4.3 8/1 0 80.0 % 0.0

Growth and Settler Survival

The relationship between growth rates and longer-term survival (indicated by right censorship ofsettlers ~9 months ofage) is shown in Figure 3.6. Although overall, settlers with higher growth rates displayed significantly higher probabilities ofsurvival, this relationship was significant only for Porites compressa and Pocillopora damicornis, but not Montipora capitata (Table 3.7).

"0 0.12 .~ "E co 0.1 "0 --- coc..c..... en § 0.08 o E ...... 0>0> .3 a. 0.06 ...... 0>en o co ~ ~ 0.04 o 0 -en '-c --~0> _ 0.02 ...... 0 ~ a. ..c "3 o o..... M. capitata P. compressa P. damicornis <.9 Species

Figure 3.6. Mean growth rates (slopes ofloglO standardized polyp increase per month) ± S.B. ofM capitata, P. compressa and P. damicornis survivors (right censored individuals) and non-censored settlers.

68 Table 3.7. a) Factorial ANOVA ofgrowth rates (slope ofloglO standardized polyp increase per month) ofM. capitata, P. compressa and P. damicornis survivors (right censored individuals) and non-censored settlers; b) Binomiallogit homogeneity ofslopes model ofsurvival (examines survival as the dependent variable and growth as an explanatory variable); c) Within species contrasts ofgrowth rates between right censored and non-censored settlers. a) Source df SS MS F p Intercept 1 0.7784 0.7784 444.9728 0.0000 Species 2 0.1718 0.0859 49.0951 0.0000 Rcen 1 0.0536 0.0536 30.6674 0.0000 Species*Rcen 2 0.0285 0.0143 8.1479 0.0004 Error 258 0.4513 0.0017 b) Source df Wald Stat. P Intercept 1.000 7.8872 0.0050 Species 2.000 5.7070 0.0576 Growth 1.000 19.0507 0.0000 Species*growth 2.000 6.3552 0.0417 c) Dead Live M capitata P. compressa P. damicornis M capitata 1.0000 P. compressa 0.0000 P. damicornis 0.0123

Settlement, Survival, Growth and Habitat

Exploratory analysis ofgeneral relationships between the means ofmeasured habitat characteristics (water motion, total suspended solids, sedimentation rates, salinity, temperature, % adult coral cover; Table 3.8) and mean site settlement, early growth and survival ofMontipora capitata, Porites compressa and Pocillopora damicornis was conducted using a Pearson's correlation matrix (Table 3.9). The only significant correlation among habitat characteristics (not shown) was noted between mean water motion and mean rates ofsedimentation (Pearson's correlation, r = 0.9057, n = 6, P =

0.013). Montipora capitata, P. compressa and P. damicornis differed in the types of significant habitat correlations they displayed, with bay water clarity (total suspended solids) and salinity appearing potentially important to M capitata settlement and survival, water motion and water clarity potentially important to P. compressa survival,

69 Table 3.8. Means (± S.E.) and ranges (in parentheses) of recorded habitat characteristics at monitored settlement sites in Kane 'ohe Bay (C = Central Bay,~= North B~ = South Bay, F = Fringing reef, P = Patch reef) Total % Adult Cover/m2 Water Suspended Sediment Motion Solids Rate Salinity Temperature 2 Site (DF) (mg/l) (mglcm day) (0%) __ _ (DC) M.~itata _ P. comJJressa P. damicornis CF 10.72 ±0.70 1.85 ± 0.22 5.25 ± 1.42 33.6 ± 0.4 26.0 ± 0.9 1.56 ± 0.43 10.21 ± 3.56 0.26 ± 0.14 (8.74-13.92) (0.86-2.64) (2.01-16.49) (31.5-35.9) (22.9-27.8) (0.65-2.45) (3.65-16.55) (0.08-0.67)

CP 13.47 ± 0.93 1.98 ± 0.29 10.36 ± 3.40 33.6 ±0.4 26.3 ±0.6 64.24 ±7.70 8.77 ± 3.84 0.15 ±0.1O (10.03-18.06) (0.97-3.75) (2.77-33.28) (31.8-36.0) (24.7-28.0) (42.73-78.44) (2.55-19.93) (0.00-0.45)

NF 19.84 ± 1.09 3.25 ± 0.68 7.54 ±0.62 33.3 ± 0.5 26.4 ± 0.8 20.22 ± 8.74 18.13±4.85 0.58 ± 0.20 (15.50-24.27) (1.42-6.62) (4.31-10.27) (31.2-36.7) (23.6-27.8) (6.50-45.78) (10.62-31.71) (0.09-0.98)

NP 22.92 ± 1.32 4.22 ± 1.89 8.20 ± 4.34 33.6 ± 0.5 25.8 ± 0.8 12.35 ± 5.00 9.21 ± 3.66 0.14 ± 0.06 (17.89-28.97) (1.23-17.16) (2.69-47.14) (31.4-37.0) (22.7-27.4) (2.00-23.55) (1.65-18.70) (0.03-0.32) o-....l SF 4.60 ± 0.36 1.92 ± 0.20 1.53 ± 0.19 33.6 ± 0.4 26.1 ± 0.8 32.26 ± 3.13 34.05 ± 4.85 0.14 ± 0.03 (3.57-6.69) (0.78-2.75) (0.70-2.63) (31.6-36.0) (23.2-28.1) (26.70-41.03) (24.70-47.53) (0.05-0.18)

SP 4.30 ± 0.37 1.71 ± 0.11 1.52 ± 0.12 33.6 ± 0.4 26.0 ±0.6 39.35 ± 7.24 44.10 ± 9.59 0.07 ±0.02 (3.16-6.12) (1.03-1.92) (0.86-1.96) (31.5-36.0) (23.8-27.7) (28.93-60.60) (17.35-62.98) (0.05-0.13) Table 3.9. Pearson's correlations matrix of means of measured habitat characteristics and mean number of settlers, settler growth and percent survival for three species across six sites in Kane'ohe Bay. Correlation coefficients and associated P-values presented. (Number of settlers m'2 and water motion log10 transformed; growth = slope ofloglO standardized polyp increase per month; % 12 month survival and % adult cover m'2 arcsine square-root transformed; total suspended solids and sediment lOglO +1 transformed).

Total Water Motion Sediment Temperature % Adult Species Suspended Salinitye%) (DF) (mg/cm2day) (0C) 2 Solids (mgll) Cover/m No. M. capitata -0.4451 -0.8671 -0.1969 0.5673 -0.0416 0.0359 Settlers/m2 0.3764 0.0253 0.7084 0.2404 0.9376 0.9462 P. compressa 0.6036 0.3353 0.4079 -0.0130 -0.4851 -0.7305 0.2046 0.5158 0.4221 0.9806 0.3295 0.0992 P. damicomis 0.8610 0.4627 0.8942 -0.4341 0.2489 0.4448 0.0276 0.3555 0.0162 0.3897 0.6344 0.3767

Growth rate M. capitata -0.3660 -0.0608 -0.4764 -0.3430 -0.3800 0.1476 0.5447 0.9227 0.4172 0.5720 0.5281 0.8128 -.l ...... P. compressa 0.7015 0.3866 0.6275 -0.2728 -0.2722 -0.6517 0.1203 0.4490 0.1823 0.6010 0.6018 0.1608 P. damicomis -0.4927 -0.4752 -0.9306 0.7746 -0.7970 -0.4045 0.5073 0.5248 0.0694 0.2254 0.2030 0.5955

% 12 month M. capitata -0.6858 -0.7269 -0.6933 0.9616 -0.6186 0.1563 Survival 0.1326 0.1017 0.1267 0.0022 0.1904 0.7675 P. compressa 0.8150 0.9820 0.5615 -0.6614 0.0235 -0.4260 0.0482 0.0005 0.2463 0.1525 0.9648 0.3996 P. damicomis 0.0072 0.2142 -0.2014 -0.6924 0.2051 0.9325 0.9928 0.7858 0.7986 0.3076 0.7949 0.0675 and water motion andlor sedimentation rate potentially important to settlement ofP. damicornis. Mean growth did not correlate significantly with means ofany ofthe measured habitat characteristics. No significant relationship between a species' mean

2 number ofsettlers m- , early growth or survival and corresponding adult distributions was found (Table 3.9).

Site Projection Example

A general projection ofsettlement and survival throughout each 500 m2 study area is shown for summer 2000 settlers ofMontipora capitata, Porites compressa and

Pocillopora damicornis in Table 3.10. Projected survivor numbers at four, 12 and 21 months were significantly positively correlated with initial settler numbers ofM. capitata

(Pearson's correlations, P < 0.05), suggesting an overall influence oflarvae availability on numbers ofpost-settlement survivors in this species. Significant positive correlations were also evident for P. compressa and P. damicornis at 4 months (Pearson's correlations, P < 0.05), but were not present at 21 months (Pearson's correlations, P >

0.05).

Interspecific variability in survival dramatically changed proportional settler representation over time at some sites (Table 3.10). Montipora capitata dominated initial counts at four ofthe six sites (67 %), but by 21 months remained the dominant representative species at one (17 %). Porites compressa, although dominant initially at only two sites (33 %), was projected to dominate survivor numbers at four ofthe six sites

(83 %) within the 21-month period. The projection ofzero survival in zones with known adult abundance (Table 3.10) suggests site-specific variation in settler survival over time.

72 Table 3.10. Projected numbers of settlers and survivors in 500 m2 zones of measurement representing six sites in Kane'ohe Bay (C =Central Bay, N =North Bay, S =South Bay, F =Fringing Reef, P =Patch Reef) Mean No. Settlers/m2 Colonizable Minimum No.of4 No. of 12 No. of 21 colonizable Substrate No. Initial Month Month Month 2 Site Species substrate (m ) Settlers Survivors Survivors Survivors CF M. capitata 21,553 286 6.168 x 106 81,999 28,700 3,417 P. compressa 338 286 96,834 38,956 24,932 11,986 P. damicornis 79 286 22,595 12,980 5,079 2,049

CP M. capitata 1,932 77 149,512 3,214 536 201 P. compressa 79 77 6,108 2,489 995 622 P. damicornis 90 77 6,981 2,493 1,247 831

NF M. capitata 4 169 637 3 0 0 P. compressa 79 169 13,376 10,744 6,208 4,356 P. damicornis 26 169 4,459 3,358 1,099 488

-.l NP M. capitata 8 163 1,227 15 3 0 VJ P. compressa 357 163 58,259 39,271 27,150 19,635 P. damicornis 218 163 35,569 15,869 3,967 1,700

SF M. capitata 338 50 16,875 271 135 90 P. compressa 94 50 4,688 1,875 852 0 P. damicornis 8 50 375

SP M. capitata 83 50 4,166 383 96 0 P. compressa 53 50 2,651 994 398 249 P. damicornis 4 50 189 Discussion

Site and Seasonal Variation

Spatial and temporal variability in coral settlement and recruitment is a phenomenon common to many reefsystems (Harrison and Wallace 1990 and cited references, Fisk and Harriott 1990, Samarco 1991, Tomascik 1991, Smith 1992, 1997,

Fitzhardinge 1993, Glynn et al. 1994, 1996,2000, Gleason 1996, Baird and Hughes 1997,

Connell et al. 1997, Dustan and Johnson 1998, Fairfull and Harriott 1999, Hughes et al.

1999,2000,2002, Carlon 2001, Kojis and Quinn 2001, Jinendradasa and Ekarantne

2003) and has been noted previously in Kane'ohe Bay (Fitzhardinge 1993). In this study, overall settlement appeared highest at Central Bay sites, particularly Central Fringing reef, although the relative significance ofsettlement at this site varied among the most common coral species. Levels ofMontipora capitata settlement at Central Bay sites were up to two orders ofmagnitude higher than South and North Bay sites. Porites compressa, however, did not differ in overall mean settlement between sites, and settlement at

Central and North Bay sites was not significantly different for Pocillopora damicornis, although both were significantly higher than South Bay sites. These differences in distributional patterns ofsettlement did not correspond to site-specific adult abundances, and may be a consequence ofdifferences in fertilization strategies (i.e., location of fertilization, see Kolinski and Cox 2003) and/or the level ofsynchronization in gamete/planulae release, in combination with water flow patterns within Kane'ohe Bay and between bay and oceanic zones (Bathen 1968, Smith et al. 1981, Figure 3.7).

At a gross level ofmeasurement, settlement over time corresponded well with observed spawning times ofthe major coral species. Montipora capitata was found to

74 settle mainly in the summer. This was the only species observed to display discrete settlement pulses soon after peaks in spawning. In contrast, P. compressa did not display discrete spawning peaks and exhibited greater temporal dispersion ofsettlement, which occurred mostly in summer and fall. Pocillopora damicornis is a year round brooder in

Hawai'i (Edmondson 1946, Harrigan 1972, Jokie11985, Jokiel et al. 1985). Although it has been suggested that significant P. damicornis settlement would be restricted to summer months in Hawaiian waters (Jokiel and Guinther 1978), such a pattern was not apparent when comparing levels ofsettlement across seasons in this study. The tendency to reproduce year round in corals is most often associated with brooding species

(Harrison and Wallace 1990), suggesting that C. tenella is a brooder.

The apparent disassociation between levels ofsettlement and adult abundances suggests some separation ofsource and sink populations within Kane'ohe Bay (Holthus et al. 1989, Fitzhardinge 1993). Central Fringing reef, for example, appeared as a sink for

M capitata and P. compressa larvae despite relatively low site abundance ofadult populations (particularly M capitata). Water flow within this area is tidal, but is also influenced by wind driven long-shore currents originating in the South Bay (Bathen

1968, Smith et al. 1981, Figure 3.7). The residency time ofSouth Bay waters is approximately 13 days (Smith et al. 1981). Embryonic development ofM capitata and P. compressa occurs within 2 to 3 days (Mate et al. 1998, Hunter 1988a, Kolinski Chapter 2 and unpublished data), and larvae ofM. capitata are highly competent to settle from 3 to

42 days post fertilization (Kolinski Chapter 2). High density adult populations at South

Bay sites, in combination with flow patterns, water residency times and timing for embryonic larval development, suggests there may be a connectivity corridor between

75 South Bay and the Central Fringing reefarea, at least for M. capitata. Further

investigation ofconnectivity within the bay using genetic analysis is currently underway

(Kinzie et al. 2003).

21°30'

o

N r

o 1 2 3 ======, km ;,'

157° 50' 157°45' Figure 3.7. Generalized seawater circulation pattern in Kane'ohe Bay. Modified from Bathen (1968) and Smith et al. (1981).

76 Settler Survival

Type III survivorship was strongly evident for Montipora capitata, and appeared

also for Porites compressa and Pocillopora damicornis settlers at Kane'ohe Bay sites.

Such a pattern ofsurvival is common to marine invertebrates (reviewed by Gosselin and

Qian 1997, Hunt and Scheib1ing 1997), including many sc1eractinian coral species for which settler survival has been examined (Babcock 1985, Sato 1985, Babcock and

Mundy 1996, Mundy and Babcock 2000, Baird and Hughes 2000, Raimondi and Morse

2000). Numerous factors may influence early survival, including genetic unsuitability or abnormalities, low energy reserves, predation, accidental ingestion, bulldozing or trampling by mobile invertebrates, competition, disease, and a variety ofabiotic disturbances (such as changes in water motion, water clarity, light exposure, sedimentation, salinity, etc.; Gosselin and Qian 1997, Hunt and Scheibling 1997). Field mortalities exceeding 99% have previously been reported for M capitata in Kane'ohe

Bay within a few months following settlement (Fitzhardinge and Bailey-Brock 1989).

Assuming significant loss ofgametes through failure to fertilize (Oliver and Willis 1987,

Oliver and Babcock 1992, Levitan 1995, Lasker et al. 1996, Underwood and Keough

2001) and high mortality ofembryos and larvae in the plankton and when settling

(Underwood and Fairweather 1989, Morgan 1995, Pechenick 1999), such repeated loss of settled individuals suggests great investment ofreproductive effort may be needed by individuals ofM capitata to increase the likelihood ofeventual replacement with fecund offspring (note: high production may occur with a relatively minor cost, Kolinski Chapter

4). Spatial and temporal variation in habitat suitability is a likely driving force towards such reproductive "waste" (Underwood and Keough 2001). The dispersal oflarvae

77 presumably "spreads the risk" (sensu Reddingius and den Boer 1970) ofencountering one or more habitats in which offspring are genetically suited to survive beyond maturity

(Williams 1975, Maynard Smith 1978, Bell 1982, Jablonski and Lutz 1983, Bull et al.

1987, Holt and McPeek 1996, Underwood and Keough 2001).

Montipora capitata settlers did not appear to be as suited for survival on shallow

Kane'ohe Bay reefs as P. compressa and P. damicornis. Initially this was assumed to be the result ofspecies differences in proportions ofvery recent settlers observed across censuses. Although this assumption eventually needs to be tested, results ofthe analysis examining proportional survival offour-month survivors at 12 and 21 months tended to support the initial findings ofsurvival differences. Variability in survival did exist among sites, and temporal variability must occur, as is evidenced by adult abundances at sites with zero settler survival. A better comparative test might involve the seeding and monitoring ofmultiple plates at multiple sites with similar aged larvae ofeach species repeatedly over a number ofyears.

The differences in survival ofM. capitata, P. compressa and P. damicornis settlers can dramatically alter settler year-class representation at a site over time. This was most evident at Central Fringing reef, where roughly 6 million projected M capitata setters ended up being outnumbered by 100,000 P. compressa settlers after 21 months had elapsed. The tendency for P. compressa to dominate settler survivorship numbers may partially explain this species success in occupying space in a number of environments. Montipora capitata success in gaining occupancy ofspace may partially depend on repetitive influxes ofvery large numbers ofsettlers. A significant positive correlation between initial numbers ofM capitata settlers and survivors over time was

78 found, and there was some suggestion that a minimum threshold density ofsettlers may

be needed to ensure long-term (21 months in this study) year class representation ofthis

species at a site.

Settler Growth

The mean standardized rate ofgrowth ofMontipora capitata was exceedingly

slow compared to Porites compressa and Pocillopora damicornis, although potential

growth rates (i.e., fastest recorded) were not too dissimilar. The three species differ in

larvae size, and thus presumably available energy at settlement. In this study, larval size had no apparent influence on mean rates ofoverall settler growth, as was evidenced by P.

compressa (which has the smallest larvae) displaying a significantly higher mean growth rate than the larger M capitata, and a statistically similar rate ofgrowth to that ofP. damicornis (largest larvae ofthe three species). Previous comparisons between broadcast

spawning and planulating species have suggested relatively faster early growth rates for planulating species (Babcock 1985, Harrison and Wallace 1990). Clearly this was not the case for P. compressa and P. damicornis in Kane'ohe Bay. Montipora capitata, P. compressa and P. damicornis all possess zooxanthellae upon parental release as eggs or

larvae. Thus, differences in mean early growth cannot be attributed to any apparent need ofany ofthese species to acquire and multiply zooxanthellae following settlement (see

Babcock 1985, Van MoorseI1988). Internal factors other than those related to initial size, available energy, and zooxanthellae acquisition ostensibly play a critical role(s) in the growth determination ofsettlers ofthese three species. Although not detected in this

study, environmental factors also likely playa very important role.

79 A large proportion (61 %) ofthe surviving M. capitata settlers displayed an

ability to remain in a "dormant" one-polyp state for extended periods oftime exceeding

one year. This phenomenon was not observed in P. compressa or P. damicornis (small

numbers ofone-polyp, year old P. compressa settlers were observed; however, such

status was associated with previous declines in polyp numbers), but may not be unique as

similar delays in growth have been observed in Goniastrea aspera in Japan (Sakai 1998)

Variable rates in growth and an ability to maintain "dormancy" across multiple breeding

seasons have implications for appropriate assessment ofage/year class representation and

correlation to observed reproductive phenomena (Hughes and Connell 1987, Muko et al.

2001). New observations ofnumerous large visible recruits ofM capitata on a reefmay

say more about a particular year's conditions affecting growth ofcumulative settlers that

arrived across a number ofyears than about any single reproductive event involving high reproductive output, fertilization success and/or levels ofsettlement. Age determinations

ofeven small M capitata recruits will be difficult to determine without direct monitoring

from the period ofinitial settlement.

A relationship between rates ofgrowth and longer-term (nine plus months)

survival was not evident for M. capitata, but was significant for P. compressa and P. damicornis. Fast early growth leading to occupation ofspace is generally considered to be advantageous in competitive and/or frequently disturbed habitats (Jackson 1977,

Birkeland 1977, Kojis and Quinn 1981, Jackson and Hughes 1985, Szmant 1986, Johnson

1992, Sakai 1998). Growth apparently is important for the persistence ofP. compressa

and P. damicornis, both ofwhich displayed temporally scattered and somewhat lower overall levels ofsettlement. Montipora capitata may gain its advantage through high

80 settler numbers occupying a variety ofspaces that, when given opportunity and

appropriate conditions, grow and recruit into the community. The comparison may thus

be one ofquality versus quantity.

A Comparison of Parameters and Suggested Recruitment Strategies

A summary ofreproduction and recruitment related findings for Montipora

capitata, Porites compressa and Pocillopora damicornis in Kane'ohe Bay are listed in

Table 3.11 as a means to understand and compare early life history strategies. Various

characteristics differ among the three species. As mentioned above, the early life history

strategy ofM. capitata appears to be heavily focused around quantity, based mainly on reproductive characteristics that lead to high levels oflarva1 development and settlement.

Discrete periods ofspawning synchrony, hermaphrodism and surface water fertilization are all characteristics that increase potential for high levels oflarval development. The larvae display a mainly generalist strategy in terms ofsettlement substrate preference

(Kolinski Chapter 2). Although M. capitata larvae remain competent to settle for extended periods oftime (Kolinski Chapter 2), the vast majority ofrecruitment occurs in pulses soon after spawning during the summer. High overall mortality ofsettlers is displayed (Fitzhardinge and Bailey-Brock 1989, Fitzhardinge 1993, this study), suggesting internal and/or external qualitative inadequacies. On average, growth is slow.

Although M capitata settlers appear to possess a potential for relatively rapid growth, such growth may be delayed until favorable conditions prevail. An ability ofsome settlers to survive in the absence ofgrowth suggests that an accumulation ofyear classes may persist and recruit to the community together when conditions allow.

81 Table 3.11. Life history characteristics ofM capitata, P. compressa and P. damicornis in Kane'ohe Bay e= Hodgson 1986; 2 = Krupp et al. 1992; 3 = Po1acheck 1978) Characteristic M. capitata P. compressa P. damicornis Sex Hermaphrodite Gonochoric Hermaphrodite

Reproductive mode Broadcast Spawner Broadcast Spawner Brooder

Reproductive Season Summer - Fall Summer - Fall Year Round Highly Synchrony in SpawninglPlanulation Extended Extended Synchronized Location ofFertilization Water Surface Water Column Polyps

Relative Larval Size Mid Size - Large Small Large

1 Peak Larval Numbers in Plankton 16000 1000 108 Tows (per 100 m3) Settlement Timing Highly Pulsed Extended Extended

Settlement Preference Generalist Unknown Generalist

Relative Mean Settler Growth Rate Slow Fast Fast Relative Potential Settler Growth Fast Very Fast Very Fast Rate Single Polyp "Dormancy" Yes No No

Settler Survivorship Type III Type III Type III

Mean One-year Survival (%) 0.4 (± 0.1 S.E.) 16.7 (±4.8 S.E.) 8.7 (± 1.3 S.E.)

Significant Correlation: No. Initial Yes No No Settlers vs. No. 21 Month Survivors

Settler Growth-Survival No Yes Yes Relationship 2Maturation Size (Colony Diameter, 9.5 (?) 3.5 (?) 2.5 cm) 2 Pheonix Effect in Settlers No Yes No

Adult Community Persistence Long-Term Long-term Short-term 3Maximum Diameter Commonly > 1 m > 1 m 12 -14 cm Observed

In contrast to M. capitata, P. compressa and P. damicornis appear to place emphasis on certain qualitative characteristics in their early life histories. In P. compressa, separate sexes, extended, less synchronized periods ofspawning, and fertilization within the three-dimensional water column potentially limit larval numbers

(see Hodgson 1986) but leads to an extended settlement period over time. Type III

82 survivorship ofsettlers exists; however, known mortality levels relative to that ofM capitata are significantly lower. High relative rates ofgrowth are evident within this species, which may be explained by the finding that survival is significantly related to growth. Several ofthe settlers, however, displayed an ability to "resurrect" from what appeared to be a state ofdeath (Kolinski in prep.). Fast early growth and high levels of survival are qualitative characteristics likely influencing the persistence ofthis species.

Pocillopora damicornis has growth and survival characteristics very similar to that ofP. compressa. However, P. damicornis in Kane'ohe Bay is a brooder, albeit with reported asexually derived larvae (Stoddart 1986), producing and releasing planulae year round

(Jokiel1985). Maximum colony size and total area coverage within Kane'ohe Bay are much lower than that ofM. capitata and P. compressa, and distributions are generally restricted to shallow water reefflats and margins (Maragos 1972, Polacheck 1978,

Maragos et al. 1985, Holthus et al. 1989, Hunter and Evans 1995). Thus, the total number oflarvae released can be expected to be much lower than that produced by M. capitata and P. compressa in a given year (see Hodgson 1986, Hunter and Evans 1995). The larvae are able to settle soon after release (Harrigan 1972) but remain competent for periods of at least 100 days (Richmond 1987a). Pocillopora damicornis larvae appear to be generalists in terms ofsettlement preference (see Kolinski Chapter 2). The settlers grow relatively rapidly and display relatively high proportional levels ofsurvival. Their early survival does appear to be linked to growth. Pocillopora damicornis has been noted as an opportunistic species (Stoddart 1983, Richmond 1988, Endean and Cameron 1990,

Jokiel1998). It tends to be the first species observed on new substrates in Kane'ohe Bay

(Fitzhardinge 1993, this study), but is noted for determinate/size limited growth

83 (Maragos, pers. comm.) and high rates ofturnover within populations (Maragos 1972,

Polacheck 1978, Richmond 1985b, 1987b, Jokiel1998). Based on reported colony sizes

at maturity (Polacheck 1978), surviving P. damicornis settlers are likely to be the first of the three species to begin gamete/larval production.

84 CHAPTER 4

ENERGY ALLOCATION TO REPRODUCTION

Introduction

Sexual reproduction in scleractinian corals involves the formation and dispersal of

planktonic larvae in habitats that vary spatially and temporally. Corals are iteroparous,

mainly modular organisms, with a potential to be reproductively prolific. The

implications oflarge, repetitive reproductive output include increased probabilities of

successful fertilization, dispersal, settlement and recruitment over time (Underwood and

Fairweather 1989, Oliver and Babcock 1992, Levitan 1995, Hughes et al. 2000). Rates of

mortality in the early life history ofcorals, however, tend to be high (Babcock 1985,

Fitzhardinge and Bailey-Brock 1989, Fitzhardinge 1993, Babcock and Mundy 1996,

Fairfull and Harriott 1999, Kolinski Chapter 3), bringing into question the magnitude and

importance ofparental investment and the costs associated with reproduction.

Lipids are energy rich molecules critical in a number ofways to various early life history strategies ofcorals. Lipids provide buoyancy to eggs-sperm bundles ofnumerous

coral species for fertilization in the two-dimensiona11ike environment ofthe waters

surface (Babcock et al. 1986, Harrison 1988, Harrison and Wallace 1990, Arai et al.

1993, Richmond 1997). Development ofcora11arvae and lengths ofcompetency and

dispersal appear to be partially dependent on the energy derived from stored lipids

(Richmond 1988, Arai et al. 1993, see also Wilson and Harrison 1998, Isomura and

Nishihira 2001), which make up roughly 70 % ofthe composition ofeggs and larvae in

species investigated (Richmond 1987a, Arai et al. 1993). Lipids may influence the

vertical migration oflarvae through the water column, with lipid expulsion being used to

85 decrease buoyancy for potential settlement substrate exploration (Rinkevich and Loya

1979, Van Moorse11983, Richmond 1987a). Larvae also secrete mucus-lipids for initial

adherence to settlement substrates (Vandermeulen 1975, Lewis 1974, Chia and Bickell

1978, Goreau et al. 1981, Carlon and Olson 1993). Presumably, the metamorphosis and

early growth ofmany settlers relies initially on remaining energy reserves oflipids

derived from parental sources.

Most hermatypic corals are restricted to environments with suitable levels of photosynthetically active radiation due to obligate symbiotic relationships with dinoflagellate zooxanthellae (Falkowski et al. 1990, Muscatine 1990, Barnes and Chalker

1990). These corals receive large quantities ofenergy translocated from zooxanthellae in lipid form (Crossland et al. 1980, Patton et al. 1983, Muscatine et al. 1984, Muscatine

1990). Lipids associated with reproductive products have been directly linked to this autotrophic source ofnutrition (Rinkevich 1989). Translocated energy is typically believed sufficient to encompass the energetic needs ofshallow water colonies on sunny days (Davies 1984, 1991, Falkowski et al. 1984, Edmunds and Davies 1986, Stimson

1987, Muscatine 1990, Rinkevich 1996). Variation in photosynthetically active radiation over time, however, may limit energy available for colony metabolism (Davies 1991).

Although disagreement exists (Rickevich 1996), energy allocation to reproduction in corals has generally been viewed in terms oftradeoffs with other colony processes, suggesting that energy, at times, may be limiting (Kojis and Quinn 1984, Rinkevich and

Loya 1985, Harrison and Wallace 1990, Kozlowski 1992, Ward 1995a, Oren et al. 1997,

2001, Kramarsky-Winter and Loya 2000).

86 Direct examination ofthe energetic costs ofreproduction in relation to energy

uptake has only occurred for two brooding species, with projections restricted to sunny

days (Edmunds and Davies 1986, Davies 1991, Jokie11998). This study examined energy

allocation to reproduction in the broadcast spawning coral Montipora capitata in relation

to energy derived from photosynthesis on an annual basis. Montipora capitata is a

hermaphroditic coral that spawns annually during summer months. It releases buoyant,

lipid-filled egg-sperm bundles that congregate and fertilize at the waters' surface

(Heyward 1986, Kolinski and Cox 2003). Although tending to settle soon after

embryogenesis, larvae ofthis species can remain competent to settle for up to seven months (Kolinski Chapter 2). Large concentrations oflarvae (Hodgson 1986) and settlers

(Kolinski Chapter 3) have been observed in the field; however, settler survival is

extremely low, even relative to other common species ofscleractinian corals (Kolinski

Chapter 3). These early life history traits ofM capitata are viewed from the perspective ofenergetic costs associated with reproduction in this species.

Methods

Collection of Corals and Measurement of Environmental Parameters

Colonies ofMontipora capitata were collected from 1 m depth on the windward reefflat ofCoconut Island and from 3 to 4 m depth offthe windward reefslope of

Kokokahi (close to the depth limits ofM capitata at this site), Kane'ohe Bay, O'ahu,

Hawai'i (Figure 4.1). The colonies were tagged and placed in cleared areas at their respective collection sites and depths four months prior to monitoring for gamete release in 1997. A total of 14 of 18 colonies collected at Coconut Island and 17 of 18 colonies at

87 N r

o 2 .....

1570 50' 1570 45' Figure 4.1. Coral collection sites in Kane'ohe Bay, O'ahu, Hawai'i. Kokokahi were successfully monitored. Maximum diameters were measured in May

prior to spawning and ranged between 15 and 20 em. Colony displacement volumes were

3 measured in August following spawning and ranged between 162 and 833 cm . All

selected colonies had greater than 95 % live tissue cover. Additional colonies ofvarious

sizes were collected from these sites and depths in 1998 to increase sample sizes for various gamete measurements as described below.

Environmental parameters were measured on a monthly basis between August

1996 and July 1997 at three locations along a 50-meter transect at each colony collection

site and depth. Light attenuation at depth was measured using two Li-Cor Inc. model LI-

188B units with LI-1925B sensors (note: measurements occurred in only nine ofthe 12 months due to equipment malfunction). Sensors were equilibrated at the water's surface,

88 and then one sensor was placed face up on the substrate at depth. Five measurements of mean photosynthetically active radiation (PAR; 400 to 700 nm) were made over 10 second intervals at each location. Mean values were used to determine mean percent PAR at depth and extinction coefficients for each month at each site. Three one-liter water samples were collected at monthly intervals in clear Nalgene™ polyethylene bottles at each site for measurement oftotal suspended solids (TSS). All water was collected 0.15 m above the substrate. The samples were vacuum filtered onto pre-weighed 500DC pre­ ashed GC50 filters, with ample rinsing using de-ionized water. The TSS samples were placed in a 60 DC drying oven for 24 hours prior to being weighed on a Mettler AB104 electronic microbalance. Rates ofsedimentation were measured for 24-hour periods at monthly intervals using 15.24 cm deep, 5.45 cm wide, bottom-capped polyvinyl chloride

(PVC) pipe traps. Two traps were clipped to PVC posts at each ofthe three locations at each site. The traps were plugged prior to removal from the posts and were vacuum filtered onto individual pre-weighed 500DC pre-ashed GC50 filters with ample rinsing using de-ionized water. The sediments were dried at 60DC for 24 hours and then weighed on a Mettler AB 104 electronic microbalance. Oxygen concentrations were measured monthly using a YSI model 55. Oxygen flux on reefs is known to increase throughout the day (Kohn and Helfrich 1957, Kinsey 1985), which complicated comparisons between sites, as measurements could not be made simultaneously at both Coconut Island and

Kokokahi. Data exploration showed oxygen concentrations at Coconut Island to be consistently higher than Kokokahi despite measurement time, thus only those values obtained when Coconut Island was measured first where used in a pair-wise analysis

(one-sided, paired by month). Salinity and temperature were measured monthly using a

89 YSI model 30 conductivity meter. Water motion was determined using Plaster ofParis

"clod cards" (Doty 1971, Jokiel and Morrissey 1993), with two clod cards being

anchored to each ofthree bricks at each site per trial for a period of24-hours. Control

cards were placed in a five-gallon (19l) bucket ofseawater in a shaded, wind protected

area on land. The cards were briefly rinsed in still fresh water, air-dried for a month (in

an air-conditioned, dehumidified room) and then weighed on a Mettler P160 balance.

The values ofeach measured parameter were averaged for each month at each site for parametric or non-parametric pair-wise comparisons.

Reproductive Output

Spawning ofMontipora capitata is a predictable event, occurring between the

months ofMay and September during new and first quarter moons between 20:45 and

22:30 (Hunter 1988b, Kolinski and Cox 2003). Colonies were transferred to flow-through

sea water tables covered with 15 % shade-cloth at the Hawai'i Institute ofMarine

Biology two to three days prior to each new moon from May through September 1997. At

18:00 each evening, colonies were placed in submerged, individually sized black plastic

flowerpots with pot upper-lips maintained slightly above the water level ofeach table

such that each pot contained the buoyant gamete bundles ofa single colony once released. Open ports in pot bottoms allowed for water exchange without loss ofgametes.

Water exchange was enhanced by the presence oflarge aerators placed in the seawater tables. The colonies were carefully returned to the tables each night after 22:45, and were returned to the field after each month's predicted/observed lunar associated spawning period. Colonies remained submerged during all transfers between field and lab and int%ut ofpots.

90 Surface waters within pots were examined for gamete presence under low, usually

red, light conditions. Gametes, upon spawning, were collected with disposable plastic

pipettes and were counted individually and/or collectively measured for volume in 10 ml

and 25 ml graduated cylinders. Volume measurements never exceeded 2.3 ml for any

sample, and all measurements occurred prior to significant bundle breakup. A colony

volume to gamete bundle number relationship (No. Bundles = 843.6*Volume - 26.3; n =

2 30, R = 0.95) was determined for colonies monitored in 1997 and 1998, using one

representative sample from each colony in regression analysis, satisfying appropriate

model assumptions. Collective measurements for each 1997 colony were scored

throughout the spawning season, with measured volumes being converted to bundle

numbers using the regression.

Energy Estimates

Gamete bundles were collected from 11 Coconut Island (five in 1997 and six in

1998) and 11 Kokokahi colonies (six in 1997 and five in 1998) and were pipetted directly

onto individual screens of 100 /lm Nitex mesh, quickly rinsed with de-ionized water to remove salts (bundles and eggs did not lyse), placed in plastic Gelman petri-dishes,

frozen at -50°C, and then lyophilized for 12 to 16 hours. Bundle samples between 8.59

and 10.88 mg from each container were molded into small balls using clean stainless

steel spatulas. Samples were weighed on an Ohaus Analytical Plus electronic balance.

Energy content was determined using a Phillipson microbomb calorimeter (see Phillipson

1964). Multiple samples from each colony were measured, and caloric contents were

calculated utilizing the mean ofthree benzoic acid standards. Additional samples were burned on pre-ashed and weighed GC50 filters at 500°C to determine ash content (Paine

91 1966, 1971). Mean bundle mass was determined for known numbers ofgamete bundles

from 18 colonies (four 1997, and five 1998 Coconut Island colonies and three 1997, and

six 1998 Kokokahi colonies) on an Ohaus Analytical Plus electronic balance.

Comparisons ofenergy content per mg gametes and ofbundle weights were made

between sites and times using factorial ANOVA, meeting all model assumptions.

Estimates ofenergy content ofthe total reproductive output ofcorals monitored in 1997 were determined by multiplying mean gamete values ofjoules per mg x mg per bundle x total bundles per colony.

Photosynthesis versus Irradiance Measurements

In November 2001, five Montipora capitata colonies each were collected from

Coconut Island and Kokokahi for measurement ofphotosynthesis versus irradiance (P-/) at the Hawai'i Institute ofMarine Biology. These colonies were to act as surrogates for estimating annual productivity in the monitored 1997 colonies. Collection sites and depths were the same as those in 1997 and 1998. Colonies were selected such that the sizes spanned the range for colonies monitored for reproductive output in 1997.

Maximum colony diameters were between 15 and 20 em, and all colonies had greater than 95 % live tissue cover. Colony size measurements included volume (water displacement), buoyant weight (see Jokiel et al. 1978) and projected area (area ofan ellipse = 7r x 0.5 length x 0.5 width).

Oxygen production ofM capitata colonies was measured over three clear days in four open-topped Plexiglas containers placed in two white flow-through seawater tables such that the containers were in unobstructed sunlight. Each container received seawater pumped directly from the Coconut Island reefflat such that water volumes were replaced

92 every two to three minutes. When oxygen measurements were being made, water was re­ circulated using Little Giant centrifugal pumps, maintaining vigorous water motion similar to that at colony collection sites. Water pumped through the flow-through seawater tables ensured temperatures similar to the reefflat environment were maintained in each container throughout the course ofthe study. Water temperatures during incubations remained at 26.02 °c (± 0.01 S.E.).

Photosynthetically active radiation (between 400 and 700 nm) was measured as

I1Einsteins m-2 sec-1 using a Li-Cor Inc. model LI-1400, with a LI-l92SB underwater sensor submerged adjacent to containers in each flow-through seawater table. Oxygen concentration was measured in mg min-1 using two YSI model 600XLM. Oxygen measurements were made in the following fashion: The YSI sensors were placed in a bucket offresh water. The external seawater source to two ofthe four containers (one in each table) was shut off, and the pumps were used to lower water in the containers to predetermined levels corresponding to colony size. The pumps were reconnected such that water in each container was re-circulated, and a YSI sensor was placed in the comer ofeach container for a period of 10 minutes (note: calibration trials found no difference in oxygen readings between units or between positions within the containers). The YSI sensors were then transferred to the fresh water bucket (salinity changes allowed for easy data recognition ofan individual trial), outside flow was returned to the two containers, and the process was repeated for the other two containers. Measurements were made in a staggered fashion between 04:30 and 16:30 on each ofthe three days. The colonies were placed in the containers with external seawater flow roughly six hours prior to first measurements to ensure colony stress would not occur. The colonies did not appear to be

93 stressed throughout the procedure. In darkness, polyps were extended, and mucus

production was minimal throughout the course ofthe experiment.

Rates ofoxygen production and consumption (mg min-I) were calculated as the

slope of20 oxygen measurements made (one every 30 seconds) over each 10-minute

period, correcting values for the total water volume. Gas exchange across the water

surface was assumed to be inconsequential due to the short incubation times and

extremely low wind stress on the measurement days in the sheltered containers (see

Jokiel and Morrissey 1986).

Hyperbolic tangent (tanh) function curves (see Jassby and Platt 1976, Chalker

1981) were fitted to the P-I data using the computer program SigmaPlot. All P-I relationships were expressed by the equation

P = (Pm-R) tanh (1/ Ik) + R

(Jokiel and Morrissey 1986), where P is the net oxygen flux between the coral and the

surrounding water, Pm is the maximum photosynthetic rate (horizontal asymptote ofthe

curve), R is respiration (determined as negative oxygen flux in darkness), I is the measured irradiance (PAR), and h is the saturation constant (irradiance at which the

initial slope ofcurve intercepts the horizontal asymptote).

Estimates of Colony Productivity

Estimates oftotal primary productivity from 1 June 1996 through 31 May 1997 were made by applying depth corrected irradiance (I) measurements averaged on an hourly basis over the relevant time period to the hyperbolic tangent functions and solving

for P. Photosynthetically active radiation has been measured every 10 seconds (averaged

for each hour) at the Hawai'i Institute ofMarine Biology for the past 20 years using a

94 Cambell Scientific data logging system with a Li-Cor model LI-190SB sensor. Irradiance

values were corrected using site and depth specific mean extinction coefficients

determined for Coconut Island reefflat and Kokokahi reefslope waters from August

1996 through July 1997 (n = 9) using two Li-Cor Inc. model LI-188B with LI-1925B

sensors. Hourly values ofmg oxygen were added to obtain integrated estimates ofnet

daily photosynthesis, which were then summed over the 365 days for each colony. Total

oxygen values were converted to energy units Ooules) using the assumption that the main

carbon fixation product is lipid and the pathway is via glyceraldehydes 3-phosphate to

glycerol (Davies 1991), or, as expressed in Jokiel (1998):

C02 + H20 + 489,000 J ~ 1/3(glyceraldehdye 3-phosphate) + O2.

Values ofannual respiration were determined by assuming oxygen flux per unit time was

equivalent during daylight and darkness (Chalker et al.1988, Falkowski et al. 1990, but

see Edmunds and Davies 1988) and across seasons. Hourly estimates ofoxygen flux determined during dark hours (negative values) were multiplied by 24 and then 365 to

obtain annual respiration values. The addition ofrespiration and net productivity values provided estimates ofgross photosynthetic productivity, which were converted to joules.

All measured P-Iparameters were assumed to remain stable throughout the applied period.

Estimates ofAnnual Energy Allocation to Sexual Reproduction

Correlations ofcolony size and gross and net colony productivity were insignificant (Pearson's correlations, P > 0.05) or inconsistent between sites for the

surrogate colonies, thus average productivity values Ooules) were used as a baseline for

estimating the percent ofannual gross and net productivity allocated to reproductive

95 products for the 1997 colonies. This procedure slightly increased the estimated percentages compared to values determined using size-standardized productivity. The estimates ofenergy content ofthe total reproductive output for each coral monitored in

1997 were divided by the site and time specific average annual gross and net energy values ofproductivity determined for the 2001 surrogate colonies as a means of estimating percent energy allocation to sexual reproduction.

Results

Environmental Parameters

Despite significantly higher TSS values at Kokokahi (mean difference = 0.93 ±

0.21 S.E. mg [1, Paired t-test, n = 12, df= 11, T = -4.34, P = 0.001), extinction

I I coefficients (CI mean = 0.37 ± 0.02 S.E. m- , KK mean = 0.37 ± 0.03 S.B. m- ) did not differ significantly between sites (Paired t-test n = 9, df= 8, T = -0.07, P = 0.947).

Percent light at sample depths, however, was significantly higher at Coconut Island

(mean = 77 ± 2 S.E. %) than Kokokahi (mean = 33 ± 3 S.E. %; Paired t-test ofarcsine square-root transformed data, n = 9, df= 8, T = 11.56, P = 0.000) due to depth differences between sites (0.7 m versus 3.1 m depth). Water motion was significantly higher on the Coconut Island reef flat (mean difference = 4.91 ± 0.55 S.E. DF, Paired t­ test, n = 12, df= 11, T = 8.91, P = 0.000). Sedimentation rates did not differ (Sign test, n

= 12, P = 0.387).

Oxygen concentrations were significantly higher at Coconut Island than Kokokahi

(Paired t-test, n = 3, df= 2, T = -15.53, P = 0.002), despite the fact that all Coconut Island values assessed were obtained earlier in the day than those at Kokokahi. Mean temperature was slightly but significantly higher at Coconut Island (mean difference =

96 0.3 ± 0.1 S.E. °c, Wilcoxon signed rank test, n = 12, P = 0.047). No difference in salinity

was observed (Wilcoxon signed rank test, n = 12, P = 0.879).

In addition to measured water quality differences between the two sites,

qualitatively, the Kokokahi site appeared degraded compared to the Coconut Island site.

Coral cover was less than 3 % at Kokokahi (greater than 30 % at Coconut Island).

Substrate at Kokokahi was mainly fine, black sediment that was anoxic just below the

surface, compared to white, medium grained carbonate reefflat sands at Coconut Island.

Filter feeding invertebrates, such as polychaetes and sponges, were observed to be more prevalent at Kokokahi.

Reproductive Output

Colony spawning in 1997 occurred in June, July and August. No spawning was noted in Mayor September. Seventy-four percent ofthe colonies spawned in both June and July, with two ofthe colonies spawning again in August. The percent ofgamete bundles released from each colony averaged 42.24 (± 0.07) % in June, 57.73 (± 0.07

S.E.) % in July and 0.003 (± 0.003 S.E.) % in August.

Although seven ofthe 31 colonies (23 %; two from Coconut Island and five from

Kokokahi) released less than 100 gamete bundles, all but one (97 %) ofthe monitored colonies released gametes. The average fecundity ofcolonies (standardized by colony

3 volume) from Coconut Island (mean = 6.19 ± 1.04 S.E. bundles cm- ) and Kokokahi

3 (mean = 5.35 ± 1.45 S.E. bundles cm- ) did not differ significantly (Two sample t-test with equal variance, n = 31, P = 0.654). The number ofgamete bundles released ranged

from 0 to 7834 and averaged 2962 (± 466 S.B.) bundles per colony. Colonies that released greater than 100 bundles (n = 24) averaged 3820 (± 473 S.B.) bundles per

97 3 3 colony. Colony volumes ranged from 162 to 833 cm (mean = 489 ± 33 S.B. cm ). A

significant relationship between colony size and fecundity was found (Table 4.1).

Table 4.1. Regression analysis ofcolony reproductive output (bundles) versus 3 colony size (cm ) Predictor Variables Coefficient Std. Error Student's T P Constant -484.159 1165.08 -0.42 0.6808 3 Colony Size (cm ) 7.05117 2.23212 3.16 0.0037

R2 Adj. R2 MSE Std. Dev. 0.2560 0.2304 5182646 2276.54

Source df SS MS F P Regression 1 5.172E+07 5. 172E+07 9.98 0.0037 Residual 29 1.503E+08 5182646 Total 30 2.020E+08

Energy Content ofReproductive Products

The average energy content ofgametes in bundles from Coconut Island (32.16 ±

1 1 1 0.32 S.B. J mg- ; 7.68 ± 0.08 S.E. cal mg- ) and Kokokahi colonies (32.17 ± 0.30 J mg- ;

1 7.68 ± 0.07 S.E. cal mg- ) did not differ significantly between sites or years (Table 4.2).

Pooled energy values ofgamete bundles averaged 32.17 ± 0.21 S.E. J mg-1 (7.68 ± 0.05

1 S.E. cal mg- ; n = 21 colonies). The average percent ash was 2.79 ± 0.15 S.B. Mean bundle weights did not differ significantly between sites or years (Table 4.3). The

estimated energetic content ofan individual gamete bundle ranged from 5.39 joules (1.28

cal) to 13.34 joules (3.19 cal), with a mean value of8.69 ± 0.44 S.E. joules (2.08 ± 0.11

S.E. cal; n = 18).

Table 4.2. Factorial ANDVA ofenergy content (joules) per mg ofCoconut Island and Kokokahi gametes collected in 1997 and 1998. Source df SS MS F P Site 1 2.316 2.3160 0.0339 0.8560 Year 1 3.314 3.3138 0.0485 0.8282 Site*Year 1 2.316 2.3160 0.0339 0.8560 Error 18 1230.598 68.367

98 Table 4.3. Factorial ANOVA ofmean bundle weights (mg) ofCoconut Island and Kokokahi gametes collected in 1997 and 1998. Source df SS MS F p Site 1 0.0009 0.0009 0.0111 0.9173 Year 1 0.0772 0.0772 0.9320 0.3496 Site*Year 1 0.0521 0.0521 0.6289 0.4401 Error 15 1.2428 0.0829

Applying the mean value ofgamete bundle energy content, the average energy value oftotal gametes released per colony during the 1997 reproductive season was

l 25,740 ± 4050 S.E. J coloni (6160 ± 969 S.E. cal colony-I), and ranged from 0 to 6.81 X

4 l 10 joules. The mean was 33,200 ± 4114 S.E. J colony-I (7950 ± 984 S.E. cal coloni ) for colonies (n = 24) releasing greater than 100 gamete bundles (range = 2940 to 6.81 x

104 joules).

Photosynthesis versus Irradiance Measurements and Estimates of Productivity

Correlations between measurements ofcolony size (projected area, volume, buoyant weight) and estimates ofIk, R, Pm, projected gross annual Ozproduction, net annual Oz production and gross and net photosynthetic productivity were not significant

(Pearson's correlations, P > 0.05) or were not consistent between groups within the measured size range. Although Pm values tended to be higher for colonies from Coconut

Island than Kokokahi, these differences were not significant at ex = 0.05 (mean difference

= 0.2206 ± 0.1131 S.B. mg Oz min-I, Paired t-test, one-sided, n = 4, df= 3, T = 1.95, P =

0.073; only corals ofsimilar sizes compared). Coconut Island colonies displayed z significantly higher Ik values (mean difference = 77.275 ± 32.110 S.E. p,E S-I m- , Paired t-test, one-sided, n = 4, df= 3, T = 2.41, P = 0.048) and rates ofrespiration (mean difference = -0.0841 mg Oz min-I, Paired t-test, one-sided, n = 4, df= 3, T = -2.74, P =

0.036) than similarly sized Kokokahi colonies. Hyperbolic tangent parameter values and projected values ofrespiration, gross and net Oz production, and gross and net

99 photosynthetic productivity at depth are shown in Table 4.4. Despite higher respiration rates and Ik values in Coconut Island colonies, differences in available light at depth led to significantly greater mean values ofprojected annual gross (Two-sample t-test, n = 10, df= 8, T = 2.44, P = 0.041) and net (Two-sample t-test for unequal variance, n = 10, df=

8, T = 3.07, P = 0.015) photosynthetic energy production for Coconut Island colonies.

Projected gross productivity from photosynthesis between June 1996 and May 1997 averaged 3.12 x 106 ± 5.12 x 105 S.B. joules (7.44 x 105 ± 1.22 x 105 S.B. cal) for colonies from Coconut Island and 1.61 x 106 ± 1.65 x 105 S.E. joules (3.84 x 105 ± 3.94 x

104 S.E. cal) for colonies from Kokokahi. Projected net productivity from photosynthesis averaged 14.64 x 105 ± 2.96 x 105 S.E. joules (3.49 x 105 ± 7.06 x 104 S.E. cal) for colonies from Coconut Island and 6.04 x 105 ± 1.55 x 105 S.B.joules (1.44 x 105 ± 3.71 x

104 S.E. cal) for colonies from Kokokahi.

Estimates ofAnnual Energy Allocation to Sexual Reproduction

Applying the mean site values for annual photosynthetic productivity determined above, colonies monitored for reproductive output invested between 0.003 and 1.95 %

(mean = 1.00 ± 0.18 S.E. %) oftheir gross productivity from photosynthesis to reproduction at Coconut Island, and between 0.00 and 4.23 % (mean = 1.32 ± 0.36 S.E.

%) at Kokokahi. Percent investment ofnet productivity was estimated to range from

0.003 and 4.16 % (mean = 2.13 ± 0.38 S.E. %) for Coconut Island colonies and 0.00 to

11.26 % (mean = 3.50 ± 0.95 S.E. %) for Kokokahi colonies. Considering only colonies that released greater than 100 gamete bundles, mean percentages ofgross and net values increased to 1.16 ± 0.16 S.E. % gross and 2.49 ± 0.34 S.E. % net for Coconut Island colonies, and 1.86 ± 0.41 S.E. % gross and 4.95 ± 1.09 S.B. % net for colonies from

100 Table 4.4. Photosynthesis versus irradiance hyperbolic tangent parameters, projected O2 production values and photosynthetic productivity estimates for Coconut Island (CI) and Kokokahi (KK) colonies between June 1996 and May 1997 at depth. Hyperbolic Tangent Parameters Projected production values at depth Net Gross 10 Net 10 Gross O2 O2 Respired O2 Productivity Productivity 1 1 1 1 1 Volume R I k Pm (mg yr- ) (mg yr- ) (mg yr- ) Uoules yr- ) Uoules yr- ) 3 2 4 4 4 6 5 Colony (cm ) (mg O2 min-I) CItE S-l m- ) (mg O2 min-I) x 10 X 10 X 10 X 10 X 10 CI-l 346 -0.1397 251.79 0.4940 12.98 7.34 5.64 1.98 8.62 CI-2 433 -0.1570 210.58 0.4899 13.80 8.25 5.55 2.11 8.49 CI-3 591 -0.2165 360.92 0.8813 20.27 11.38 8.89 3.10 13.58 CI-4 834 -0.2957 312.01 1.3109 31.06 15.54 15.52 4.75 23.71 CI-5 1138 -0.2221 389.51 1.1114 23.97 11.67 12.29 3.66 18.79 Mean 668 -0.2062 304.96 0.8575 20.42 10.84 9.58 3.12 14.64 KK-1 232 -0.0938 108.05 0.2476 6.94 4.93 2.01 1.06 3.08 KK-2 437 -0.1296 138.85 0.5087 12.09 6.81 5.28 1.85 8.07 KK-3 498 -0.1522 194.53 0.3916 9.10 8.00 1.10 1.39 1.68 ..... KK-4 608 -0.1174 259.02 0.7446 12.57 6.17 6.40 1.92 9.77 .....0 KK-5 804 -0.1317 320.28 0.7927 11.91 6.92 4.98 1.82 7.62 Mean 516 -0.1249 204.14 0.5370 10.52 6.57 3.95 1.61 6.04 Kokokahi. No significant differences between sites were found for any ofthe percent

allocation estimates (all comparisons using Analysis ofCovariance ofarcsine square-root

transformed percentages with colony size as the covariate, P »0.05). All estimates

apply only to colonies within the measured size range.

Discussion

Site Comparisons

Results ofthe comparative analyses ofparameter estimates derived from the

hyperbolic tangent functions for Coconut Island and Kokokahi colonies were consistent

with findings ofother depth and shade related studies involving corals (Wethey and

Porter 1976, Davies 1980, Zvalinskii et al. 1980, Falkowski and Dubinsky. 1981, Chalker

et al. 1983, Porter et al. 1984, Haramaty et al. 1997, Villinski 2003), including Montipora

capitata (Kinzie and Hunter 1987). Although collection sites differed in depth by only

two to three meters, the difference in irradiance was large and significant due to the relatively large extinction coefficients. The P-Iparameters differed in a manner that

suggested divergence in photo-adaptive responses to site-specific irradiances (see Jokiel

1988, Chalker et al. 1988). Respiration and h were significantly higher for the shallow

Coconut Island colonies. Pm values also tended to be higher for the Coconut Island

colonies. Although the latter difference was not significant at a = 0.05 (P = 0.074), small

sample size may be reason. Gross and net primary productivity ofcolonies within the

measured size range were significantly lower at Kokokahi, the deeper, more shaded site.

Overall, the comparisons ofP-Iparameters met general expectations for colonies in

different light regimes, which supported the methodology.

102 The lack ofdifference between mean percent energy allocation to reproduction between corals from the two sites is interesting given significantly different mean values ofprimary productivity and the lack of a significant difference in mean number of gamete bundles released. Reproduction, however, was highly variable for colonies at both sites, thus affecting the statistical comparison. The range in estimates ofpercent energy allocation was much greater for Kokokahi colonies, reaching a maximum of 11.26 % (of net primary productivity) compared to 4.26 % for Coconut Island colonies. Percent energy allocation to reproduction was nearly twice as high for Kokokahi than Coconut

Island colonies when considering only those colonies that released greater than 100 bundles. Site-specific variation in colony fecundities may result from a number offactors including differences in colony size, morphology, age, reproductive history, genetics, and microhabitat (see Harrison and Wallace 1990, Hall and Hughes 1996, Villinski 2003).

Tests ofhomogeneity ofvariances between sites suggest site-specific variation in fecundity is a feature common to this species.

Comparisons Between Species and Colony Processes

Mean percent energy allocation to reproduction in Montipora capitata is similar to that reported for Pocillopora damicornis (4.1 % ofnet, and 1.6 to 3 % ofgross primary production in 10 em diameter colonies; Davies 1991, Jokiel1998), but is larger than the estimate for Porites porites (0.8 % ofnet, and 0.4 % of gross primary productivity in small nubbins; Edmunds and Davies 1986). Although these values seem extremely low, some perspective may be gained through comparison with estimates ofrelative energy investments made to other colony processes, such as growth, mucus release and storage.

Davies (1991) estimated a partial energy budget for M. capitata and P. damicornis using

103 small nubbins from the Coconut Island reefslope. Growth was measured over seven

"normal" days (i.e., neither cloudless or overcast). The tissue energy content Davies

(1991) used in these calculations was that for Pocillopora eydouxi (18.05 J mg-t, Davies

1984). I substituted the value of27.5 J mg- 1 for M capitata and 25.4 J mg-1 for P.

damicornis based on measurements ofcolony tissue from Coconut Island and Kokokahi

sites (Kolinski, unpublished data; the P. damicornis value came from Coconut Island

colonies and was adjusted down 5 % to account for the potential presence ofreproductive

products, Stimson 1987, Davies 1991, Ward 1995a and b). The growth allocation

estimates for "normal" days were recalculated as 21 % for M. capitata and 37 % for P.

damicornis, or roughly 6 to 10 times the mean percent annual energy allocation to

reproduction in these species. Percent net energy allocation to growth was 14 % for P. porites (Edmunds and Davies 1986), roughly 16 times that allocated to reproduction

(based on sunny day estimates).

Although studies for a few corals suggest that large percentages ofexcess

photosynthetically fixed energy are released as mucus and dissolved organic carbon-lipid

(Crossland et al. 1980, Muscatine et al. 1984, Davies 1984, 1991, Falkowski et al. 1984,

Edmunds and Davies 1986, Crossland 1987), direct measurements are few. Crossland et

al. (1980) found 40 % ofthe 14C bicarbonate incorporated into Acropora acuminata to be

released as mucus and mucus-lipid over a 24-hour period. Mucus and dissolved organic

carbon-lipid released from Stylophora pistil/ata represented roughly 20 % ofthe

photosynthetically fixed carbon, and overall levels ofrelease were similar for Acropora

variabilis (Crossland 1987). No measurements ofenergy lost in the form ofmucus or

dissolved organic lipid have been made for M capitata, Pocillopora damicornis or

104 Porites porites; however, M. capitata and P. damicornis have been observed to release

large amounts ofmucus when stressed (Kolinski, pers. obs.). Stimson (1987, referring to

Ducklow and Mitchell 1979) suggested that mucus may be released to attract bacteria for

ingestion by coral colonies. Others suggest that shallow water corals are just getting rid

ofexcess photosynthate (Davies 1984, Falkowski et al. 1984, Muscatine 1990). Based on

data from Crossland (1987), Davies (1991) projected that the minimum excess energy released as mucus and/or dissolved organic carbon-lipid on any given day from M

capitata and P. damicornis would be 47 % ofthat remaining on an ideal day after

budgeting for colony respiration and growth. This translates to 48 % ofnet energy from primary productivity for M capitata and 58 % for P. damicornis on a "normal" day, or roughly 14 to 23 times the mean percent energy allocation to reproduction in these

speCIes.

Montipora capitata stores large amounts oflipid relative to other common

Hawaiian sc1eractinian corals (Stimson 1987, Davies 1991). Although no direct comparison ofpre- and post-spawning lipid levels has been made for M capitata,

Stimson (1987) and Davies (1991) showed tissue composition to be above 40 % lipid by weight, with a roughly five percent difference between samples from spawning and non­

spawning months (Stimson 1987). Similarly, the lipid composition in P. damicornis runs from 30 to 40 % by weight, with a roughly 5 % reduction following planulation (Stimson

1987, Davies 1991, Ward 1995a and b). Davies (1991) estimated lipid reserves could support M. capitata colonies for 114 overcast days during which respiration exceeded productivity. In P. damicornis, the estimate was 28 days. Appropriate data were not available for P. porites. Many corals appear to have the ability to reabsorb their

105 reproductive products (Rinkevich and Loya 1979, Szmant-Froelich et al. 1980, Harriott

1983, Wyers 1985, Kojis 1986, Szmant and Gassmann 1990, Harrison and Wallace 1990,

Glynn et al. 1994, Bassim 1998); thus, gametes and planulae are in a sense a secondary

storage source ofenergy for coral colonies, at least for some period oftime during

gamete formation. In Hawaiian M capitata, summertime release ofgametes correlates

with rising and peak annual levels ofirradiance; thus, energy lost from reproductive

"storage" can quickly be replaced. Estimates for M capitata based on 1997 depth

irradiance measurements suggest it would take approximately 14 July days to replace the

average energy ofreleased gamete bundles for measured Coconut Island colonies, and 30

days for Kokokahi colonies after compensating for colony respiration, 21 % energy

allocation to growth and 48 % expenditure as released mucus. These estimates increase

by month post-spawning, averaging 15 and 33 August days for Coconut Island and

Kokokahi colonies, and 18 and 39 September days for site colonies respectively. Thus,

average energy lost to reproductive release could be replaced within two to three weeks

for the shallow Coconut Island colonies, and four to six weeks for the colonies at

Kokokahi. This advantage to reproduction in summer months deserves further attention.

Limitations of Methodology

There are a number ofdifferent methodologies that can be applied to estimate

energy allocation to reproduction in corals (Richmond 1987b, Edmunds and Davies 1986,

Jokiell998, Rose and Bradley 1998, Leuzinger et al. 2003, Villinski 2003, this study).

None, however, is without limitations, as certain components ofthe coral energy budget

are presently difficult to quantify. At least two factors in this study potentially bias

percent reproductive allocation estimates upwards. These include a lack ofaccounting for

106 heterotrophic sources ofenergy (see Coma et al. 1998, Anthony and Fabricius 2000) and

evidence that daylight respiration rates may be higher than those determined in darkness

(this underestimates gross productivity and thus overestimates percent ofgross energy

allocation to reproduction, Edmunds and Davies 1988). In contrast, other factors may have led to underestimating percent reproductive energy allocation. There was no

accounting for the energy used in gamete formation or energy associated with materials

lost and utilized in repair ofsomatic tissue following spawning (Leuzinger et al. 2003).

The effects ofchange in colony size, morphology, and thus, photosynthetic efficiency

(Jokie! and Morrissey 1986) over time have not been accounted for in any studies. In addition, Pocillopora damicornis and Porites porites percent reproductive allocation

estimates were restricted to productivity values for clear, sunny days (Edmunds and

Davies 1986, JokielI998). Extrapolation to longer time periods requires compensating for irradiance fluctuations due to variable cloud cover (Davies 1991). The extent to which these factors offset each other in each ofthese species is unknown, but estimates of percent energy allocation to reproduction would remain low.

Application ofthe P-Imodels for M capitata did not incorporate the potential for seasonal variation in P-Iparameters. Respiration, Ik and Pm vary as a function oflight

(Wethey and Porter 1976, Davies 1980, Zvalinskii et al. 1980, Falkowski and Dubinsky

1981, Chalker et al. 1983, Porter et al. 1984, Kinzie and Hunter 1987, Haramaty et al.

1997, Villinski 2003), which varies seasonally. Monthly irradiance in November 2001 was higher than that for each month from June 1996 through March 1997. Use ofthe

November derived models thus, may have overestimated colony values ofrespiration, Ik and Pill for 10 ofthe 12 months considered. Overestimation ofrespiration and IK would

107 lead to conservative values ofannual net productivity. High Pm values would overestimate productivity. Although M capitata are known to respond to 80% differences in light availability through shifts in Pm (Kinzie and Hunter 1987), the Pm estimates for colonies naturally exposed to irradiances differing by roughly 45 % at

Coconut Island and Kokokahi did not differ significantly. Monthly irradiance from June

1996 through May 1997 ranged from -15 to + 11 % that ofNovember 2001. Application ofthe November 2001 P-Imodels may thus, overall, have resulted in conservative estimates ofannual productivity.

Conclusions

Although Montipora capitata can be reproductively prolific, this study suggests the overall average energetic cost ofreproduction to colonies within the measured colony size range is minimal. This makes sense for a species with high fertilization and settlement potential but very low settler survivorship (Kolinski Chapter 3). Large repetitive reproductive efforts at little cost increase the probability ofeventually producing reproductive offspring while minimizing the impact ofreproduction on the long-term survivorship ofparent colonies. Other factors, such as nitrogen, phosphorus and/or stem cells may limit overall fecundity in coral colonies that divert resources to preset hierarchical processes (Harrison and Wallace 1990, Rinkevich 1996). The costs related to these factors, in terms ofreproduction, have yet to be determined for any coral species. However, energy in shallow water environments does not, in general, appear limiting.

108 SUMMARY

Sexual reproduction and the early life history ofthe sc1eractinian coral, Montipora

capitata, was assessed in Kane'ohe Bay, O'ahu, Hawai'i. Montipora capitata is the only

species in Kane'ohe Bay that gives rise to large, very obvious, summertime surface-water

slicks ofgametes and developing embryos on nights and mornings following

synchronized spawning. Spawning occurs between 20:45 and 22:30 hrs during new and

first quarter moons between May and September, with the majority ofegg-sperm bundles

being released in June and/or July. Gamete bundles float to the water's surface where

fertilization takes place. Fertilization success is likely affected by the genotypic diversity

and densities ofspawning adults.

The larvae develop within 2.5 to 3 days, and are competent to settle 3 days

following spawning. Ability to settle remains high (83 % proportional settlement) at six weeks post spawning, but becomes reduced (:::;11 % proportional settlement) at 56 plus

days. A few surviving larvae may remain competent to settle for 207 plus days. Patterns

ofsettlement in the laboratory and field suggest major settlement pulses within days of

summer spawning ifsubstrate is encountered, with minor settlement occurring through

fall and winter months. Water circulation likely determines larval retention, transport and

settlement patterns. High population densities ofM. capitata in bays, inlets and along naturally protected shorelines, enhance probabilities offertilization and may be critical to

long-term species persistence in more exposed areas by providing sources oflarvae.

Long-term competency to settle may partially explain the presence ofM capitata across the Hawaiian Archipelago and possibly throughout surrounding regions.

109 Montipora capitata larvae are stimulated to settle and metamorphose by external,

substrate associated, cues. The larvae discriminate between substrates for settlement, but

display overall generalist settlement tendencies in that they settle equally well on a variety ofdifferent substrates including various crustose coralline algae species and

filamentous algae covered rubble. This finding corresponds with the high tolerance of

adults for a wide range ofhabitat conditions. Treatments with antibiotics suggest bacteria playa role in inducing M capitata larval settlement. Gamma aminobutyric acid did not induce settlement. The level ofsubstrate conditioning required by M capitata larvae for settlement is reached relatively rapidly, suggesting M capitata may be among the early

settlers on newly created, disturbed or introduced substrates.

Larvae ofthree common shallow water coral species, Montipora capitata (a hermaphroditic broadcast spawner), Porites compressa (a gonochoric broadcast spawner) and Pocillopora damicornis (a brooder) settled in high enough numbers on settlement plates to allow monitoring and comparison oftheir early life history attributes. Montipora capitata accounted for 91 % ofroughly 20,000 observed settlers at six sites across

Kane'ohe Bay over a two-year period. Average annual settler numbers were greatest at monitored Central Bay sites, and reached 2.8 x 104 settlers m-2 for M. capitata (Central

Fringing reef), compared to 788 settlers m-2 for P. compressa (Central Fringing reef), and

680 settlers m-2 for P. damicornis (North Fringing reef). Type III survivorship was evident for M. capitata, P. compressa and P. damicornis, but survival was significantly lower for M. capitata (mean 12-month post-settlement survivorship < 1 %, compared to

17 % for P. compressa and 9 % for P. damicornis). Although significant for P. compressa and P. damicornis, a relationship between settler growth and survival was not

110 evident for M. capitata. The mean estimated time for settlers to reach 1 cm2 projected

area in the field was 4.9 years for M capitata and 1.7 years for P. compressa and P.

damicornis. A large proportion (61 %) ofsurviving M capitata settlers displayed an

ability to remain in a "dormant" one-polyp state for extended periods oftime exceeding

one year. This phenomenon was not observed in the other species. These patterns of

settlement, survival and early growth in Montipora capitata suggest a reproductive

strategy based on larval quantity as opposed to quality, more so than observed in the

other two species.

The apparent need for large reproductive output given high settler mortality

brought into question the magnitude and importance ofparental investment and costs

associated with reproduction in Montipora capitata. This was assessed for colonies from two sites in Kane'ohe Bay in terms ofenergetic costs. Average fecundity in colonies at

Coconut island and Kokokahi did not differ despite a roughly 45 % difference in

available irradiance between these sites. Fecundity was related to colony size. The mean percent ofannual net photosynthetic productivity allocated to reproduction was estimated to be 2.13 % for Coconut Island colonies and 3.50 % for Kokokahi colonies. These values are low compared to estimates of21 % energy allocation to growth and a minimum 48 % energy allocation to mucus production. The summertime release of

gametes in M capitata correlates with rising and peak annual levels ofirradiance. It was

estimated the energy released in gamete bundles could be replaced within two to three weeks for Coconut Island colonies and four to six weeks for colonies at Kokokahi when

considering growth and mucus production.

111 Montipora capitata, a broadcast spawning sclenlctinian coral, may gain its early life history advantage through large repetitive, low cost, influxes oflarvae settling in a variety ofhabitats. When given opportunity and appropriate conditions, the collection of multiple year-class survivors, although relatively few, grow and recruit into the community. Such a strategy puts emphasis on high adult survivorship and reproductive output. Adult M capitata have a high tolerance for survival in a variety ofhabitats and frequently attain sizes greater than 1 m. A susceptibility to fragmentation (data collected but not addressed in this study) offers a mechanism that may augment genotypic longevity, allowing for localized dispersal that may enhance fertilization success amongst new neighbors.

The results ofthis investigation are applicable to emerging management issues including coral reefconnectivity, recovery and restoration. Montipora capitata has great potential for reefconnectivity within and among archipelagos. However, large repetitive inputs oflarvae may be needed for successful recruitment. Larval settlers may not be visible for years due to low survival, slow growth and growth "dormancy". Efforts should be made to determine whether certain habitat factors accelerate settler growth and increase survival. Although M. capitata is not a good candidate for larval seeding efforts in reefrestoration, it is a species tolerant ofa wide range ofhabitat conditions. Methods ofreefrestoration involving adult transplantation and fragment seeding are more suitable for this species (data collected but not addressed in this study).

112 LITERATURE CITED

Abe, N. 1937. Post-larval development ofthe coral Fungia actiniformis var. palawensis

Doderlein. Palao Tropical Biological Station Studies 1: 73- 93.

Andrefouet, S., P. J. Mumby, M. McField, C. Hu and F. E. Muller-Karger. 2002.

Revisiting coral reefconnectivity. Coral Reefs 21: 43 - 48.

Anthony, K. R. Nand K. E. Fabricius. 2000. Shifting roles ofheterotrophy and

autotrophy in coral energetics under varying turbidity. Journal ofExperimental

Marine Biology and Ecology 252: 221 - 253.

Arai, T., M. Kato, A. Heyward, Y. Ikeda, T. Iizuka and T. Maruyama. 1993. Lipid

composition ofpositively buoyant eggs ofreefbuilding corals. Coral Reefs 12: 71

-75.

Atoda, K. 1953. The larva and postlarval development ofthe reef-building corals.

Science Reports ofthe Tohoku University, Ser. 4, 20: 105 - 121.

Avila, c., C. T. Tamse and A. M Kuzirian. 1996. Induction ofmetamorphosis in

Hermissenda crassicornis larvae (Mollusca: Nudibranchia) by GABA, choline

and serotonin. Invertebrate Reproduction and Development 29: 127 - 141.

Ayre, D. J. and T. P. Hughes. 2000. Genotypic diversity and gene flow in brooding and

spawning corals along the Great Barrier Reef, Australia. Evolution 54: 1590­

1605.

Ayre, D. J. and J. M. Resing. 1986. Sexual and asexual production ofplanulae in reef

corals. Marine Biology 90: 187 - 190.

113 Babcock, R. C. 1985. Growth and mortality in juvenile corals (Goniastrea, Platygyra and

Acropora): the first year. Proceedings ofthe Fifth International Coral Reef

Symposium 4: 355 - 360.

Babcock, R. C. 1988. Fine-scale spatial and temporal patterns in coral settlement.

Proceedings ofthe 6th International Coral Reef Symposium 2: 635 - 639.

Babcock, R. C. and P. Davies. 1991. Effects ofsedimentation on settlement ofAcropora

millepora. Coral Reefs 9: 205 - 208.

Babcock, R. C. and A. 1. Heyward. 1986. Larval development ofcertain gamete­

spawning sc1eractinian corals. Coral Reefs 5: 111-116.

Babcock R. C. and C. Mundy. 1996. Coral recruitment: consequences ofsettlement

choice for early growth and survivorship in two sc1eractinians. Journal of

Experimental Marine Biology and Ecology 206: 179 - 201.

Babcock, R. C., G. D. Bull, P. L. Harrison, A. J. Heyward, J. K. Oliver, C. C. Wallace

and B. L. Willis. 1986. Synchronous spawnings of 105 sc1eractinian coral species

on the Great Barrier Reef. Marine Biology 90: 379 - 394.

Baird, A. 1998. The length ofthe larval phase in corals: new insights into patterns of

coral connectivity. Australian Coral ReefSociety Autumn Newsletter, June 1998:

6-8.

Baird, A. H. and T. P. Hughes. 1997. Spatial variation in coral recruitment around Lizard

Island, Australia. Proceedings ofthe 8th International Coral ReefSymposium 2:

1207 - 1210.

114 Baird, A. H. and T. P. Hughes. 2000. Competitive dominance by tabular corals: an

experimental analysis ofrecruitment and survival ofunderstorey assemblages.

Journal ofExperimental Marine Biology and Ecology 251: 117 - 132.

Banks, S. A. and V. J. Harriott. 1996. Patterns ofcoral recruitment at the Gneering

Shoals, southeast Queensland, Australia. Coral Reefs 15: 225 - 230.

Barlow, L. A. 1990. Electrophysiological and behavioral responses oflarvae ofthe red

abalone (Haliotis rufescens) to settlement-inducing substances. Bulletin ofMarine

Science 46: 537 - 554.

Barnes, D. J. and B. E. Chalker. 1990. Calcification and photosynthesis in reef-building

corals and algae. Pages 109 - 131 in Z. Dubinsky, ed. Ecosystems ofthe world,

coral reefs. Elsevier Science Publishers, B.V., Amsterdam.

Bassim, K. M. 1998. Effects oftemperature-induced "bleaching" on reproductive output

in Montipora verrucosa Vaughan (Scleractinia). Pages 53 - 59 in E. F. Cox, D. A.

Krupp and P. L. Jokiel, eds. Reproduction in reefcorals: results ofthe 1997

Edwin W. Pauley summer program in marine biology. Hawai'i Institute ofMarine

Biology Technical Report No. 42.

Bathen, K. H. 1968. A descriptive study ofthe physical oceanography ofKaneohe Bay,

Oahu, Hawaii. Hawai'i Institute ofMarine Biology Technical Report 14. 353 pp.

Begon, M., J. L. Harper and C. R. Townsend. 1986. Ecology: individuals, populations

and communities. Sinauer Associates, Inc. Publishers, Sunderland. 876 pp.

Bell, G. 1982. The masterpiece ofnature: the evolution and genetics ofsexuality.

University ofCalifornia Press, Berkeley. 635 pp.

115 Bell, J. D. and R. Galzin. 1984. Influence oflive coral cover on coral-reeffish

communities. Marine Ecology Progress Series 15: 265 - 274.

Benzie, J. A. H. 1999. Genetic structure ofcoral reeforganisms: ghosts ofdispersal past.

American Zoologist 39: 131 - 145.

Birkeland, C. 1977. The importance ofrate ofbiomass accumulation in early

successional stages ofbenthic communities to the survival ofcoral recruits.

Proceedings ofthe Third International Coral ReefSymposium 1: 15 - 21.

Boidron-Metairon,1. F. 1995. Larval Nutrition. Pages 223 - 248 in L. McEdwards, ed.

Marine invertebrate larvae. CRC Press, Boca Raton.

Brown, B. E. 1997. Disturbances to reefs in recent times. Pages 354 - 379 in C.

Birkeland, ed. Life and death ofcoral reefs. Chapman & Hall, New York.

Brown, B. E. and L. S. Howard. 1985. Assessing the effects of"stress" on reefcorals.

Advances in Marine Biology 22: 1 - 63.

Bruno, J. F. and M. D. Bertness. 2001. Habitat modification and facilitation in benthic

marine communities. Pages 201 - 218 in M. D. Bertness, S. D. Gaines and M. E.

Hay, eds. Marine community ecology. Sinauer Associates, Inc., Sunderland.

Bull, J. J., C. Thompson, D. Dg and R. Moore. 1987. A model for natural selection of

genetic migration. The American Naturalist 129: 143 - 157.

Caley, M. J., M. H. Carr, M. A. Hixon, T. P. Hughes, G. P. Jones and B. A. Menge. 1996.

Recruitment and the local dynamics ofopen marine populations. Annual Review

ofEcology and Systematics 27: 477 - 500.

116 Carleton, J. H. and P. w. Sammarco. 1987. Effects of substratum irregularity on success

ofcoral settlement: quantification by comparative geomorphological techniques.

Bulletin ofMarine Science 40: 85 - 98.

Carlon, D. B. 2001. Depth-related patterns ofcoral recruitment and cryptic suspension­

feeding invertebrates on Guana Island, British Virgin Islands. Bulletin ofMarine

Science 68: 525 - 541.

Carlon, D. B. and R. R. Olson. 1993. Larval dispersal distance as an explanation for adult

spatial pattern in two Caribbean reefcorals. Journal ofExperimental Marine

Biology and Ecology 173: 247 - 263.

Chalker, B. E. 1981. Simulating light-saturation curves for photosynthesis and

calcification by reef-building corals. Marine Biology 63: 135 - 141.

Chalker, B. E., W. C. Dunlap and J. K. Oliver. 1983. Bathymetric adaptations ofreef­

building corals at Davies Reef, Great Barrier Reef, Australia. II. Light saturation

curves for photosynthesis and respiration. Journal ofExperimental Marine

Biology and Ecology 73: 37 - 56.

Chalker, B. E., D. J. Barnes, W. C. Dunlap and P. L. Jokie!. 1988. Light and reef-building

corals. Interdisciplinary Science Review 13: 222 - 237.

Chia, F. and L. R. Bickell. 1978. Mechanisms oflarval attachment and the induction of

settlement and metamorphosis in coelenterates: a review. Pages 1 - 12 in F. Chia

and M. E. Rice, eds. Settlement and metamorphosis ofmarine invertebrate larvae.

Elsevier, New York.

117 Chiappone, M. and K. M. S. Sealey. 2000. Marine reserve design criteria and measures of

success: lessons learned from the Exuma Cays Land and Sea Park, Bahamas.

Bulletin ofMarine Science 66: 691 - 705.

Coffroth, M. A. and H. R Lasker. 1998. Larval paternity and male reproductive success

ofa broadcast-spawning gorgonian, Plexaura kuna. Marine Biology 131: 329-

337.

Coma, R, M. Ribes, J Gi1i and M. Zabala. 1998. An energetic approach to the study of

life-history traits oftwo modular colonial benthic invertebrates. Marine Ecology

Progress Series 162: 89 - 103.

Connell, J. H. 1973. Population ecology ofreef-building corals. Pages 205 - 245 in O. A.

Jones and R Endean, eds. Biology and geology ofcoral reefs. Vol. II. Academic

Press, New York.

Connell, J. H. 1978. Diversity in tropical rain forests and coral reefs. Science 199: 1302­

1310.

Connell, J. H. 1997. Disturbance and recovery ofcoral assemblages. Proceedings ofthe

8th International Coral ReefSymposium 1: 9 - 22.

Connell, J. H. and M. J. Keough. 1985. Distrubance and patch dynamics ofsubtidal

marine animals on hard substrata. Pages 125 - 151 in S. T. A. Pickett and P. S.

White, eds. The ecology ofnatural disturbances and patch dynamics. Academic

Press Inc., Orlando.

Connell, J. H., T. P. Hughes and C. C. Wallace. 1997. A 30-year study ofcoral

abundance, recruitment, and disturbance at several scales in space and time.

Ecological Monographs 67: 461 - 488.

118 Crisp, 1974. Factors influencing the settlement ofmarine invertebrate larvae. Pages 177­

265 in P. T. Grant and A. M. Mackie, eds. Chemoreception in marine organisms.

Academic Press, New York.

Crossland, C. J. 1987. In situ release ofmucus and DOC-lipid from the corals Acropora

variabilis and Stylophora pistil/ata in different light regimes. Coral Reefs 6: 35 ­

42.

Crossland, C. J., D. J. Barnes and M. A. Borowitzka. 1980. Diurnal lipid and mucus

production in the staghorn coral Acropora acuminata. Marine Biology 60: 81 ­

90.

Crowder, L. B., S. J. Lyman, W. F. Figueira and J. Priddy. 2000. Source-sink population

dynamics and the problem ofsiting marine reserves. Bulletin ofMarine Science

66: 799 - 820.

Dana, T. F. 1975. Development ofcontempory eastern Pacific coral reefs. Marine

Biology 33: 355 - 374.

Davies, P. S. 1980. Respiration in some Atlantic reefcorals in relation to vertical

distribution and growth form. Biological Bulletin 158: 187 - 194.

Davies, P. S. 1984. The role ofzooxanthellae in the nutritional energy requirements of

Pocillopora eydouxi. Coral Reefs 2: 181- 186.

Davies, P. S. 1991. Effect ofdaylight variations on the energy budgets ofshallow-water

corals. Marine Biology 108: 137 - 144.

Dayton, P. K., E. Sala, M. J. Tegner and S. Thrush. 2000. Marine reserves: parks,

baselines and fishery enhancement. Bulletin ofMarine Science 66: 617 - 634.

119 Deemers, P. 1996. The edge effect on a submerged coral reef: sedimentation as a

controlling factor for coral survival, growth and recruitment. M.S. Thesis,

Department ofOceanography, University ofHawai'i. 96 pp.

Deevey, E. S. 1947. Life tables for natural populations ofanimals. Quarterly Review of

Biology 22: 283 - 314.

Dollar, S. J. 1975. Zonation ofreefcorals offthe Kona coast ofHawaii. M.S. Thesis,

Department ofOceanography, University ofHawai'i. 183 pp.

Doty, M. S. 1971. Measurement ofwater movement in reference to benthic algal growth.

Botanica Marina 14: 32 - 35.

Ducklow, H. W. and R. Mitchell. 1979. Bacterial populations and adaptations in the

mucus layers on living corals. Limnology and Oceanography 24: 715 -725.

Duerden, J. E. 1904. The coral Siderastrea radians and its postlarval development.

Carnegie Institute ofWashington Publications 20: 1 - 130.

Dustan, P. K. and C. R. Johnson. 1998. Spatio-temporal variation in coral recruitment at

different scales on Heron Reef, southern Great Barrier Reef. Coral Reefs 17: 71 -

81.

Economou, A. N. 1991. Is dispersal offish eggs, embryos and larvae an insurance against

density dependence? Environmental Biology ofFishes 31: 313 - 321.

Edmondson, C. H. 1946. Behavior ofcoral planulae under altered saline and thermal

conditions. Occasional Papers ofthe Bernice P. Bishop Museum 18: 283 - 304.

Edmunds, P. J. and P. S. Davies. 1986. An energy budget for Porites porites

(Scleractinia). Marine Biology 92: 339 - 347.

120 Edmunds, P. J. and P. S. Davies. 1988. Post-illumination stimulation ofrespiration rate in

the coral Porites porites. Coral Reefs 7: 7 - 9.

Emlet, R. B. 1991. Functional constraints on the evolution oflarval forms ofmarine

invertebrates: experimental and comparative evidence. American Zoologist 31 :

707 -725,

Endean, R. and A. M. Cameron. 1990. Trends and new perspectives in coral-reef

ecology. Pages 469 - 492 in Z. Dubinsky, ed. Ecosystems ofthe world, coral

reefs. Elsevier Science Publishers B. V., Amsterdam.

English, S., C. Wilkinson and V. Baker (eds.). 1997. Survey manual for tropical marine

resources. 2nd Edition, Australian Institute ofMarine Science, Townsville. 390 pp.

Fadlallah, Y. H. 1983. Sexual reproduction, development and larval biology in

sc1eractinian corals, a review. Coral Reefs 2: 129 - 150.

Fairfull, S. J. L. and V. J. Harriott. 1999. Succession, space and coral recruitment in a

subtropical fouling community. Marine and Freshwater Research 50: 235 - 242.

Falkowski, P. G. and Z. Dubinsky. 1981. Light-shade adaptation ofStylophora pistillata,

a hermatypic coral from the GulfofEilat. Nature 289: 172 - 174.

Falkowski, , P. G., P. L. Jokiel and R. A. Kinzie III. 1990. Irradiance and corals. Pages 89

- 107 in Z. Dubinsky, ed. Ecosystems ofthe world, coral reefs. Elsevier Science

Publishers B.V., Amsterdam.

Falkowski, P. G., Z. Dubinsky, L. Muscatine and J. W. Porter. 1984. Light and

bioenergetics ofa symbiotic coral. Bioscience 34: 705 - 709.

121 Fisk, D. A. and V. J. Harriott. 1990. Spatial and temporal variation in coral recruitment

on the Great Barrier Reef: implications for dispersal hypotheses. Marine Biology

107: 485 - 490.

Fitzhardinge, R. C. 1993. The ecology ofjuvenile Hawaiian corals. Ph.D. Dissertation,

Department ofZoology, University ofHawai'i. 252 pp.

Fitzhardinge, R. C. and J. H. Bailey-Brock. 1989. Colonization ofartificial reefmaterials

by corals and other sessile organisms. Bulletin ofMarine Science 44: 567 - 579.

Fleck, J. 1997. Phosphatidylinositol (PI) signaling and subsequent events in

metamorphosis induction ofcnidarian larvae. Proceedings ofthe 8th International

Coral ReefSymposium 2: 1225 - 1230.

Fong, P. and D. Linnan. 1995. Hurricanes cause population expansion ofthe branching

coral Acropora palmata (Scleractinia): wound healing and growth patterns of

asexual recruits. P.S.Z.N. I: Marine Ecology 16: 317 - 335.

Gaines, S. D. and K. D. Lafferty. 1995. Modeling the dynamics ofmarine species: the

importance ofincorporating larval dispersal. Pages 389 - 412 in L. McEdward,

ed. Marine invertebrate larvae. CRC Press, Boca Raton.

Gardner, T. A, 1. M. Cote, J. A Gill, A Grant and A R. Watkinson. 2003. Long-tenn

region-wide declines in Caribbean corals. Science 301: 958 - 960.

Gilmour, J. 1999. Experimental investigation into the effects ofsuspended sediment on

fertilization, larval survival and settlement in a scleractinian coral. Marine

Biology 135: 451 - 462.

122 Gleason, M. G. 1996. Coral recruitment in Moorea, French Polynesia: the importance of

patch type and temporal variation. Journal ofExperimental Marine Biology and

Ecology 207: 79 - 101.

Glynn, P. W., S. B. Colley, J. H. Ting, J. L. Mate and H. M. Guzman. 2000. Reefcoral

reproduction in the eastern Pacific: Costa Rica, Panama, and Galapagos Islands

(Ecuador). IV. Agariciidae, recruitment and recovery ofPavona varians and

Pavona sp.a. Marine Biology 136: 785 - 805.

Glynn, P. W., S. B. Colley, N. J. Gassman, K. Black, J. Cortes and J. L. Mate. 1996. Reef

coral reproduction in the eastern Pacific: Costa Rica, Panama, and Galapagos

Islands (Ecuador). III. Agariciidae (Pavona gigantean and Gardineroseris

planulata). Marine Biology 125: 579 - 601.

Glynn, P. W., S. B. Colley, C. M. Eakin, D. B. Smith, J. Cortes, N. J. Gassman, H. M.

Guzman, J. B. Del Rosario and J. S. Feingold. 1994. Reefcoral reproduction in

the eastern Pacific: Costa Rica, Panama, and Galapagos Islands (Ecuador). II.

Poritidae. Marine Biology 118: 191 - 208.

Golbuu, y. 1996. Substratum preferences in coral planula larvae. M.S. Thesis,

Department ofBiology, University ofGuam. 38 pp.

Goreau, N. I., T. J. Goreau and R. L. Hayes. 1981. Settling, survivorship and spatial

aggregation in planulae and juveniles ofthe coral Porites porites (Pallas).

Bulleting ofMarine Science 31: 424 - 435.

Gosselin, L. A. and P. Y. Qian. 1997. Juvenile mortality in benthic marine invertebrates.

Marine Ecology Progress Series 146: 265 - 282.

Grigg, R. W. 1981. Acropora in Hawaii. II. Zoogeography. Pacific Science 25: 15 - 24.

123 Grigg, R. W. 1983. Community structure, succession and development ofcoral reefs in

Hawaii. Marine Ecology Progress Series 11: 1-14.

Grigg, R. W. 1997. Paleoceanography ofcoral reefs in the Hawaiian-Emperor chain

revisited. Coral Reefs 16: S33 - S38.

Grigg, R. W. and S. J. Dollar. 1990. Natural and anthropogenic disturbance on coral

reefs. Pages 439 - 452 in Z. Dubinsky, ed. Ecosystems ofthe world, coral reefs.

Elsevier Science Publishers B. V., Amsterdam.

Grigg, R. W. and R. Hey. 1992. Paleoceanography ofthe tropical eastern Pacific ocean.

Science 255: 172 - 178.

Grigg, R. W. and J. E. Maragos. 1974. Recolonization ofhermatypic corals on

submerdged lava flows in Hawaii. Ecology 55: 387 - 395.

Hall, V. R. and T. P. Hughes. 1996. Reproductive strategies ofmodular organisms:

comparative studies ofreef-building corals. Ecology 77: 950 - 963.

Haramaty, L, Y. Achituv and Z. Dubinsky. 1997. Morphology, photoadaptation and

autotrophy in hermatypic corals. Proceedings ofthe 8th International Coral Reef

Symposium 1: 855 - 860.

Harii, S., M. Omori, H. Yamakawa and Y. Koike. 2001. Sexual reproduction and larval

settlement ofthe zooxanthellate coral Alveoporajaponica Eguchi at high

latitudes. Coral Reefs 20: 19 - 23.

Harper, J. L. 1977. Population biology ofplants. Academic Press, London. 892 pp.

Harrigan, 1. F. 1972. The planulae larva ofPocillopora damicornis: lunar periodicity of

swarming and substratum selection behavior. Ph.D. Dissertation, Department of

Zoology, University ofHawai'i. 213 pp.

124 Harriott, V. J. 1983. Reproductive ecology offour scleractinian species at Lizard Island,

Great Barrier Reef. Coral Reefs 2: 9 - 18.

Harriott, V. J. 1992. Recruitment patterns ofscleractinian corals in an isolated sub­

tropical reefsystem. Coral Reefs 11: 215 - 219.

Harriott, V. J. 1999. Coral recruitment at a high latitude Pacific site: a comparison with

Atlantic reefs. Bulletin ofMarine Science 65: 881 - 891.

Harriott, V. 1. and S. A. Banks. 1995. Recruitment ofscleractinian corals in the Solitary

Island Marine Reserve, a high latitude coral-dominated community in eastern

Australia. Marine Ecology Progress Series 123: 155 - 161.

Harrison, P. L. 1988. Pseudo-gynodioecy: an unusual breeding system in the scleractinian

coral Galaxeafascicularis. Proceedings ofthe 6th International Coral Reef

Symposium 2: 699 - 705.

Harrison, P. L. and C. C. Wallace. 1990. Reproduction, dispersal and recruitment of

scleractinian corals. Pages 133 - 207 in Z. Dubinsky, ed. Ecosystems ofthe

world, coral reefs. Elsevier Science Publishers RV., Amsterdam.

Hayashibara, T., S. Ohike and Y. Kakinuma. 1997. Embryonic and larval development

and planula metamorphosis offour gamete-spawning Acropora (Anthozoa,

Scleractinia). Proceedings ofthe 8th International Coral Reef Symposium 2: 1231

-1236.

Henning, G., D. K. Hofmann and Y. Benayahu. 1996. The phorbol ester TPA induces

metamorphosis in Red Sea coral planulae (Cnidaria: Anthozoa). Experientia 52:

744 -749.

125 Heslinga, G. and A. Hillmann. 1981. Hatchery culture ofthe commercial top snail

Trochus niloticus in Palau, Carolina Islands. Aquaculture 22: 35 - 43.

Heyward, A. J. 1986. Sexual reproduction in five species ofthe coral Montipora. Pages

170 - 178 in P. L. Jokiel, R. H. Richmond and R. A. Rogers, eds. Coral reef

population biology. Hawai'i Institute ofMarine Biology Technical Memorandum

No. 37; SeaGrant Cooperative Report UNIHI-SEAGRANT-CR-80-01.

Heyward, A. J. and R. C. Babcock. 1986. Self- and cross-fertilization in sc1eractinian

corals. Marine Biology 90: 191 - 195.

Heyward, A. J. and A. P. Negri. 1999. Natural inducers for coral larval metamorphosis.

Coral Reefs 18: 273 - 279.

Highsmith, R. C. 1982. Reproduction by fragmentation in corals. Marine Ecology

Progress Series 7: 207 - 226.

Hodgson, G. 1986. Preliminary observations on the abundance and distribution of

planktonic coral larvae in Kaneohe Bay, Oahu, Hawaii. Pages 206 - 233 in P. L.

Jokiel, R. H. Richmond and R. A. Rogers, eds. Coral reefpopulation biology.

Hawai'i Institute ofMarine Biology Technical Memorandum No. 37; SeaGrant

Cooperative Report UNIHI-SEAGRANT-CR-80-01.

Hodgson, G. 1990. Sediment and the settlement oflarvae ofthe reefcoral Pocillopora

damicornis. Coral Reefs 9: 41 - 43.

Hoegh-Guldberg, O. 1999. Climate change, coral bleaching and the future ofthe world's

coral reefs. Marine and Freshwater Research 8: 839 - 866.

Hoegh-Guldberg, O. and J. S. Pearse. 1995. Temperature, food availability, and the

development ofmarine invertebrate larvae. American Zoologist 35: 415 - 425.

126 Holmes, N. J., V. J. Harriott and S. A. Banks. 1997. Latitudinal variation in patterns of

colonization ofcryptic calcareous marine organisms. Marine Ecology Progress

Series 155: 103 - 113.

Holt, R. D. and M. A. McPeek. 1996. Chaotic population dynamics favors the evolution

ofdispersal. The American Naturalist 148: 709 -718.

Holthus, P. F., J. E. Maragos and C. W. Evans. 1989. Coral reefrecovery subsequent to

the freshwater kill of 1965 in Kaneohe Bay, Oahu, Hawaii. Pacific Science 43:

122 -134.

Hubbell, S. P. 1997. A unified theory ofbiogeography and relative species abundance

and its application to tropical rain forests and coral reefs. Coral Reefs 16: S9 ­

S21.

Hughes, T. P. 1985. Life histories and population dynamics ofearly successional corals.

Proceedings ofthe Fifth International Coral Reef Symposium 4: 101 -106.

Hughes, T. P. and J. H. Connell. 1987. Population dynamics based on size or age? A reef­

coral analysis. The American Naturalist 129: 818 - 829.

Hughes, T. P. and J. H. Connell. 1999. Multiple stressors on coral reefs: a long-term

perspective. Limnology and Oceanography 44: 932 - 940.

Hughes, T. P. and J. B. C. Jackson. 1985. Population dynamics and life histories of

foliaceous corals. Ecological Monographs 55: 141 - 166.

Hughes, T. P. and J. E. Tanner. 2000. Recruitment failure, life histories, and long-term

decline ofCaribbean corals. Ecology 81: 2250 - 2263.

127 Hughes, T. P., A. H. Baird, E. A. Dinsdale, N. A. Moltschaniwskyj, M. S. Pratchett, J. E.

Tanner and B. L. Willis. 1999. Patterns ofrecruitment and abundance ofcorals

along the Great Barrier Reef. Nature 397: 59 - 63.

Hughes, T. P., A. H. Baird, E. A. Dinsdale, N. A. Moltschaniwkdyj, M. S. Pratchett, J. E.

Tanner and B. L. Willis. 2000. Supply-side ecology works both ways: the link

between benthic adults, fecundity, and larval recruits. Ecology 81: 2241 - 2249.

Hughes, T. P., A. H. Baird, E. A. Dinsdale, V. J. Harriott, N. A. Moltschaniwskyj, M. S.

Pratchett, J. E. Tanner and B. L. Willis. 2002. Detecting regional variation using

meta-analysis and large-scale sampling: latitudinal patterns in recruitment.

Ecology 83: 436 - 451.

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly, C. Folke, R.

Grosberg, O. Hoegh-Guldberg, J. B. C. Jackson, J. Kleypas, J. M. Lough, P.

Marshall, M. Nystrom, S. R. Palumbi, J. M. Pandolfi, B. Rosen and J.

Roughgarden. 2003. Climate change, human impacts, and the resilience ofcoral

reefs. Science 301: 929 - 933.

Hunt, H. L. and R. E. Scheibling. 1997. Role ofearlypost-settlment mortality in

recruitment ofbenthic marine invertebrates. Marine Ecology Progress Series 155:

269 - 301.

Hunte, W. and M. Wittenberg. 1992. Effects ofeutrophication and sedimentation on

juvenile corals. II. Settlement. Marine Biology 114: 625 - 631.

Hunter, C. L. 1988a. Genotypic diversity and population structure ofthe Hawaiian reef

coral, Porites compressa. Ph.D. Dissertation, Department ofZoology, University

ofHawai'i. 136 pp.

128 Hunter, C. L. 1988b. Environmental cues controlling spawning in two Hawaiian corals,

Montipora verrucosa and M. dilitata. Proceedings ofthe 6th International Coral

ReefSymposium 2: 727-732.

Hunter, C. L. 1993. Genotypic variation and clonal structure in coral populations with

different disturbance histories. Evolution 47: 1213 - 1228.

Hunter, C. L. and C. W. Evans. 1995. Coral reefs in Kaneohe Bay, Hawaii: two centuries

ofwestern influence and two decades ofdata. Bulletin ofMarine Science 57: 501

- 515.

Imhoff, J. F. 1986. Osmoregulation and compatible solutes in eubacteria. FEMS

Microbiology Reviews 39: 57 - 66.

Isomura, N. and M. Nishihira. 2001. Size variation ofplanulae and its effect on the

lifetime ofplanulae in three pocilloporid corals. Coral Reefs 20: 309 - 315.

Jablonski, D. 1995. Extinctions in the fossil record. Pages 25-44 in J. H. Lawton and R.

M. May, eds. Extinction rates. Oxford University Press, Oxford.

Jablonski, D. and R. A. Lutz. 1983. Larval ecology ofmarine benthic invertebrates:

paleobiological implications. Biological Reviews 58: 21- 89.

Jackson, J. B. C. 1977. Competition on marine hard substrata: the adaptive significance

ofsolitary and colonial strategies. The American Naturalist 111: 743 - 767.

Jackson, J. B. C. 1995. Constancy and change oflife in the sea. Pages 45 - 54 in 1. H.

Lawton and R. M. May, eds. Extinction rates. Oxford University Press, Oxford.

Jackson, J. B. C. and T. P. Hughes. 1985. Adaptive strategies ofcoral-reefinvertebrates.

American Scientist 73: 265 - 274.

129 Jackson, G. A. and R. R. Strathmann. 1981. Larval mortality from offshore mixing as a

link between precompetent and competent periods ofdevelopment. The American

Naturalist 118: 16 - 26.

Jassby, A. D. and T. Platt. 1976. Mathematical formulation ofthe relationship between

photosynthesis and light for phytoplankton. Limnology and Oceanography 21 :

540 - 547.

Jinendradasa, S. S. and S. U. K. Ekaratne. 2003. Post-bleaching changes in coral

settlement at the Hikkaduwa Nature Reserve in Sri Lanka. Proceedings ofthe 9th

International Coral ReefSymposium 1: 417 - 420.

Johannes, R. E. 1975. Pollution and degradation ofcoral reefcommunities. Pages 13 - 51

in E. J. Ferguson-Wood and J. E. Johannes, eds. Tropical marine pollution.

Elsevier Oceanographic Series 12, Elsevier Scientific Publishers, New York.

Johannes, R. E. 1978. Reproductive strategies ofcoastal marine fishes in the tropics.

Environmental Biology ofFishes 3: 65 - 84.

Johnson, C. R. and D. C. Sutton. 1994. Bacteria on the surface ofcrustose corallint algae

induce metamorphosis ofthe crown-of-thorns starfish Acanthasterplanci. Marine

Biology 120: 305 - 310.

Johnson, C. R., D. G. Muir and A. L. Reysenback. 1991. Characteristic bacteria

associated with surfaces ofcoralline algae: a hypothesis for bacterial induction of

marine invertebrate larvae. Marine Ecology Progress Series 74: 281 - 294.

Johnson, C. R., T. E. Lewis, D. S. Nichols and B. M. Degnan. 1997. Bacterial induction

ofsettlement and metamorphosis in marine invertebrates. Proceedings ofthe 8th

International Coral ReefSymposium 2: 1219 - 1224.

130 Johnson, K. G. 1992. Population dynamics ofa free-living coral: recruitment, growth and

survivorship ofManicina areolata (Linnaeus) on the Caribbean coast ofPanama.

Journal ofExperimental Marine Biology and Ecology 164: 171 - 191.

Jokiel, P. L. 1984. Long distance dispersal ofreefcorals by rafting. Coral Reefs 3: 113 ­

116.

Jokiel, P. L. 1985. Lunar periodicity ofplanula release in the reefcoral Pocillopora

damicornis in relation to various environmental factors. Proceedings ofthe 5th

International Coral ReefSymposium Congress 4: 307 - 312.

Jokiel, P. L. 1988. Is photoadaptation a critical process in the development, function and

maintenance ofreefcommunities? Proceedings ofthe 6th International Coral Reef

Symposium 1: 187 -192.

Jokiel, P. L. 1990a. Long-distance dispersal by rafting: reemergence ofan old hypothesis.

Endeavour 14: 66 -73.

Jokiel, P. L. 1990b. Transport ofreefcorals into the Great Barrier Reef. Nature 347: 665

-667.

Jokiel, P. L. 1998. Energetic cost ofreproduction in the coral Pocillopora damicornis: a

synthesis ofpublished data. Pages 41 - 44 in E. F. Cox, D. A. Krupp and P. L.

Jokiel, eds. Reproduction in reefcorals: results ofthe 1997 Edwin W. Pauley

summer program in marine biology. Hawai'i Institute ofMarine Biology

Technical Report No. 42.

Jokiel, P. L. and S. L. Coles. 1990. Response ofHawaiian and other Indo-Pacific corals

to elevated temperature. Coral Reefs 8: 155 - 162.

131 Jokie1, P. L. and E. B. Guinther. 1978. Effects oftemperature on reproduction in the

hermatypic coral Pocillopora damicornis. Bulletin ofMarine Science 28: 786­

789.

Jokie1, P. L. and F. J. Martinelli. 1992. The vortex model ofcoral reefbiogeography.

Journal ofBiogeography 19: 449 - 458.

Jokie1, P. L. and J. 1. Morrissey. 1986. Influence ofsize on primary production in the reef

coral Pocillopora damicornis and the macroalgae Acanthophora spicijera. Marine

Biology 91: 15 - 26.

Jokie1, P. L. and J. 1. Morrissey. 1993. Water motion on coral reefs: evaluation ofthe

'clod card' technique. Marine Ecology Progress Series 93: 175-181.

Jokie1, P. L. and W. A. Tyler III. 1993. Distribution ofstony corals in Johnston Atoll

lagoon. Proceedings ofthe Seventh International Coral ReefSymposium 2: 683 ­

692.

Jokie!, P. L, R. Y. Ito and P. M. Liu. 1985. Night irradiance and synchronization oflunar

release ofplanula larvae in the reefcoral Pocillopora damicornis. Marine Biology

88: 167 - 174.

Jokie1, P. L., J. E. Maragos and L. Franzisket. 1978. Coral growth: buoyant weight

technique. Pages 529 - 541 in D. R. Stoddart and R. E. Johannes, eds. Coral reefs:

research methods. UNESCO, Paris.

Jokie1, P. L., C. L. Hunter, S. Taguchi and L. Watarai. 1993. Ecological impacts ofa

fresh-water "reefkill" in Kaneohe Bay, Oahu, Hawaii. Coral Reefs 12: 177 - 184.

132 Jokiel, P. L., E. K. Brown, A. Friedlander, S. K. Rodgers and W. Smith. 2001. Hawaii

coral reefinitiative coral reef assessment and monitoring program (CRAMP) final

report 1999-2000. Prepared for the Hawai'i Coral Reef Initiative, University of

Hawai'i and the National Ocean Service, NOAA. 90 pp.

Karlson, R. H. 1999. Dynamics ofcoral communities. Kluwer Academic Publishers,

London. 248 pp.

Kaspar, H. F. and D. O. Mountfort. 1995. Microbial production and degradation of"/­

aminobutyric acid (GABA) in the abalone larval settlement habitat. FEMS

Microbiology Ecology 17: 205 - 212.

Kinsey, D. W. 1985. Metabolism, calcification and carbon production. I. System level

studies. Proceedings ofthe Fifth International Coral Reef Congress 4: 505 - 526.

Kinzie III, R A. 1999. Sex, symbiosis and coral reef communities. American Zoologist

39: 80 - 91.

Kinzie, R A. III and T. Hunter. 1987. Effect oflight quality on photosynthesis ofthe reef

coral Montipora verrucosa. Marine Biology 94: 95 - 109.

Kinzie, RA., E. F. Cox, T. D. Lewis and S. P. Kolinski. 2003. Investigations of

connectivity among coral communities using population genetic analysis. Hawai'i

Coral ReefInitiative-RP proposal. 35 pp.

Knowlton, N. and J. B. C. Jackson. 1993. Inbreeding and outbreeding in marine

invertebrates. Pages 200 - 274 in N. W. Thornhill, ed. The natural history of

inbreeding and outbreeding, theoretical and empirical perspectives. The

University ofChicago Press, Chicago.

133 Kohn, A. J. and P. Helfrich. 1957. Primary organic productivity ofa Hawaiian coral reef.

Limnology and Oceanography 2: 241 - 251.

Kojis, B. L. 1986. Sexual reproduction in Acropora (Isopora) species (Coelenterata:

Scleractinia). I. A. cuneata and A. palifera on Heron Island reef, Great Barrier

Reef. Marine Biology 91: 291 - 309.

Kojis, B. L. and N. J. Quinn. 1981. Aspects ofsexual reproduction and larval

development in the shallow water hermatypic coral, Goniastrea australensis

(Edwards and Haime 1857). Bulletin ofMarine Science 31: 558 - 573.

Kojis, B. L. and N. J. Quinn. 1984. Seasonal and depth variation in fecundity ofAcropora

palifera at two reefs in Papua New Guinea. Coral Reefs 3: 165 - 172.

Kojis, B. L. and N. J. Quinn. 2001. The importance ofregional differences in hard coral

recruitment rates for determining the need for coral restoration. Bulletin ofMarine

Science 69: 967 - 974.

Kolinski, S. P. and E. F. Cox. 2003. An update on the modes and timing ofgamete and

planula release in Hawaiian scleractinian corals with implications for

conservation and management. Pacific Science 57: 17 - 27.

Kolinski, S. P. and P. L. Jokiel (Compilers). 2002.The development, implementation and

summary findings ofthe Hawaii Coral ReefAssessment and Monitoring Program:

three-year summary, 2002. Report prepared for the Hawai'i Coral ReefInitiative,

212 pp.

Kozlowski, J. 1992. Optimal allocation ofresources to growth and reproduction:

implications for age and size at maturity. Trends in Ecology and Evolution 7: 15 ­

19.

134 Kramarsky-Winter, E. and Y. Loya. 2000. Tissue regeneration in the coral Fungia

granulosa: the effect ofextrinsic and intrinsic factors. Marine Biology 137: 867 ­

873.

Krupp, D. A, P. L. Jokiel and T. S. Chartrand. 1992. Asexual reproduction by the

solitary scleractinian coral Fungia scutaria on dead parent coralla in Kaneohe

Bay, Oahu, Hawaiian Islands. Proceedings ofthe i h International Coral Reef

Symposium 1: 527 - 534.

Lasker, H. R. 1991. Population growth ofa gorgonian coral: equilibrium and non­

equilibrium sensitivity to changes in life history traits. Oecologia 86: 503 - 509.

Lasker, H. R. and M. A Coffroth. 1999. Responses ofclonal reeftaxa to environmental

change. American Zoologist 39: 92 - 103.

Lasker, H. R., D. A Brazeau, J. Calderon, M. A. Coffroth, R. Coma and K. Kim. 1996.

In situ rates offertilization among broadcast spawning gorgonian corals.

Biological Bulletin 190: 45 - 55.

Lee, E. T. 1992. Statistical methods for survival data analysis. John Wiley & Sons, Inc.,

New York. 482 pp.

Leitz, T. 1997. Induction ofsettlement and metamorphosis ofcnidarian larvae: signals

and signal transduction. Invertebrate Reproduction and Development 31: 109­

122.

Leuzinger, S., K. R. N. Anthony and B. L. Willis. 2003. Reproductive energy investment

in corals: scaling with module size. Oecologia 136: 524 - 531.

135 Levin, L. A. and T. S. Bridges. 1995. Pattern and diversity in reproduction and

development. Pages 1 - 48 in L. McEdward, ed. Marine invertebrate larvae. CRC

Press, Boca Raton.

Levitan, D. R. 1995. The ecology offertilization in free-spawning invertebrates. Pages

123 - 156 in L. McEdward, ed. Ecology ofmarine invertebrate larvae. CRC

Press, New York.

Lewis, J. B. 1974. The settlement behaviour ofplanulae larvae ofthe hermatypic coral

Faviafragum (Esper). Journal ofExperimental Marine Biology and Ecology 15:

165 -172.

Maida, M., J. C. ColI and P. W. Sammarco. 1994. Shedding new light on scleractinian

coral recruitment. Journal ofExperimental Marine Biology and Ecology 180: 189

-202.

Man, A., R. Law and N. V. C. Polunin. 1995. Role ofmarine reserves in recruitment to

reef fisheries: a metapopulation model. Biological Conservation 71: 197 - 204.

Maragos, J. E. 1972. A study ofthe ecology ofHawaiian reefcorals. Ph.D. Dissertation,

Department ofOceanography, University ofHawai'i. 290 pp.

Maragos, J.E. 1977. Order Scleractinia stony corals. Pages 158 - 241 in D. M. Devaney

and L. G. Eldredge, eds. Reef and shore fauna ofHawaii. Section 1: Protozoa

through Ctenophora. Bernice P. Bishop Museum Special Publication 64.

Maragos, J. E. 1993. Impact ofcoastal construction on coral reefs in the U.S.-affiliated

Pacific Islands. Coastal Management 21: 235 - 269.

136 Maragos, J. E. 1995. Revised checklist ofextant shallow-water stony coral species from

Hawaii (Cnidaria: Anthozoa: Scleractinia). Bishop Museum Occasional Papers

No. 42: 54 - 55.

Maragos, J. E. and P. L. Jokiel. 1986. Reefcorals ofJohnston Atoll: one ofthe world's

most isolated reefs. Coral Reefs 4: 141 - 150.

Maragos, J. E., C. Evans and P. Holthus. 1985. Reefcorals in Kaneohe Bay six years

before and after termination ofsewage discharges (Oahu, Hawaiian Archipelago).

Proceedings ofthe Fifth International Coral ReefCongress 4: 189 - 194.

Maragos, J. E., D. C. Potts, G. Aeby, D. Gulko, J. Kenyon, D. Siciliano and D.

VanRavenswaay. 2004. 2000-2002 rapid ecological assessment ofcorals

(Anthozoa) on shallow reefs ofthe Northwestern Hawaiian Islands. Part 1:

species and distribution. Pacific Science 58: 211 - 230, in press.

Mate, J. L., J. Wilson, S. Field and E. G. Neves. 1998. Fertilization dynamics and larval

development ofthe scleractinian coral Montipora verrucosa in Hawai'i. Pages 27

- 39 in E. F. Cox, D. A. Krupp and P. L. Jokiel, eds. Reproduction in reefcorals:

results ofthe 1997 Edwin W. Pauley summer program in marine biology. Hawai'i

Institute ofMarine Biology Technical Report No. 42.

Maynard Smith, J. 1978. Evolution ofsex. Cambridge University Press, New York. 222

pp.

Mayr, E. 1970. Populations, species, and evolution. Harvard University Press,

Cambridge, Massachusettes. 453 pp.

McCauley, D. W. 1997. Serotonin plays an early role in the metamorphosis ofthe

hydrozoan Phialidium gregarium. Developmental Biology 190: 229 - 240.

137 Mileikovsky, S. A. 1971. Types oflarval development in marine bottom invertebrates,

their distribution and ecological significance: a re-evaluation. Marine Biology 10:

193 - 213.

Miller, K. and C. Mundy. 2003. Rapid settlement in broadcast spawning corals:

implications for larval dispersal. Coral Reefs 22: 99 - 106.

Moore, D. S. and G. P. McCabe. 1993. Introduction to the practice ofstatistics. 2nd

Edition, W. H. Freeman and Company, New York. 854 p.

Morgan, S. G. 1995. Life and death in the plankton: larval mortality and adaptation.

Pages 279 - 321 in L. McEdward, ed. Marine invertebrate larvae. CRC Press,

Boca Raton.

Morse, A. N. C. 1992. Role ofalgae in the recruitment ofmarine invertebrate larvae.

Pages 385 - 403 in D. M. John, S. J. Hawkins and J. H. Price, eds. Plant-animal

interactions in the marine benthos. Systematics Association Special Volume No.

46, Clarendon Press, Oxford.

Morse, A. N. C. and D. E. Morse. 1996. Flypapers for coral and other planktonic larvae.

Bioscience 46: 254 - 262.

Morse, A. N. C., K. Iwao, M. Baba, K. Shimoike, R. Hayashibara and M. Omori. 1996.

An ancient chemosensory mechanism brings new life to coral reefs. Biological

Bulletin 191: 149 -154.

Morse, D. E. 1990. Recent progress in larval settlement and metamorphosis: closing the

gaps between molecular biology and ecology. Bulletin ofMarine Science 46: 465

-483.

138 Morse, D. E. and A. N. C. Morse. 1991. Enzymatic characterization ofthe morphogen

recognized by Agaricia humilis (sc1eractinian coral) larvae. Biological Bulletin

181: 104-122.

Morse, D. E., N. Hooker, H. Duncan and L. Jensen. 1979. ),-aminobutyric acid, a

neurotransmitter, induces planktonic abalone larvae to settle and begin

metamorphosis. Science 204: 407 - 410.

Morse, D. E., N. Hooker, A. N. C. Morse and R. A. Jensen. 1988. Control oflarva1

metamorphosis and recruitment in sympatric agariciid corals. Journal ofMarine

Biology and Ecology 116: 193 - 217.

Morse, D. E., A. N. C. Morse, P. T. Raimondi and N. Hooker. 1994. Morphogen-based

chemical flypaper for Agaricia humilis cora11arvae. Biological Bulletin 186: 172

-181.

Mountfort, D. O. and V. Pybus. 1992a. Effect ofpH, temperature and salinity on the

production ofgamma aminobutyric acid (GABA) from amines by marine

bacteria. FEMS Microbiology Ecology 101: 237 - 244.

Mountfort, D. O. and V. Pybus. 1992b. Regulatory influences on the production of

gamma aminobutyric acid (GABA) by a marine pseudomonad. Applied and

Environmental Microbiology 58: 237 - 242.

Muko, S., K. Sakai and Y. Iwasa. 2001. Size distribution for a marine sessile organism

with space-limitation in growth and recruitment: application to a coral population.

Journal ofAnimal Ecology 70: 579 - 589.

Mumby, P. J. 1999. Can Caribbean coral populations be modeled at metapopu1ation

scales? Marine Ecology Progress Series 180: 275 - 288.

139 Mundy, C. N. and R. C. Babcock. 1998. Role oflight intensity and spectral quality in

coral settlement: implications for depth-dependent settlement? Journal ofMarine

Biology and Ecology 223: 235 - 255.

Mundy, C. and R. Babcock. 2000. Are vertical distribution patterns ofsc1eractinian corals

maintained by pre- or post-settlement processes? A case study ofthree contrasting

species. Marine Ecology Progress Series 198: 109 - 199.

Muscatine, L. 1990. The role ofsymbiotic algae in carbon and energy flux in reefcorals.

Pages 75 - 87 in Z. Dubinsky, ed. Ecosystems ofthe world, coral reefs. Elsevier

Science Publishers, B.Y., Amsterdam.

Muscatine, L., P. G. Falkowski, J. W. Porter and Z. Dubinsky. 1984. Fate of

photosynthetically fixed carbon in light- and shade-adapted colonies ofthe

symbiotic coral Stylophora pistillata. Proceedings ofthe Royal Society of

London, Series B 222: 181 - 202.

Nadeau, L., J. A. Paige, B. Starczak, T. Capo, J. Lafler and J. P. Bidwell. 1989.

Metamorphic competence in Aplysia californica Cooper. Journal ofExperimental

Marine Biology and Ecology 131: 171- 193.

Negri, A. P., N. S. Webster, R. T. Hill and A. J. Heyward. 2001. Metamorphosis of

broadcast spawning corals in response to bacteria isolated from crustose algae.

Marine Ecology Progress Series 223: 121 - 131.

Nozawa, Y. and P. L. Harrison. 2003. Larval settlement patterns, dispersal potential, and

the effect oftemperature on settlement oflarvae ofthe reefcoral, Platygyra

daedalea, from the Great Barrier Reef. Proceedings ofthe 9th International Coral

ReefSymposium 1: 409 - 415

140 Ogden, J. C. 1997. Marine managers look upstream for connections. Science 278: 1414­

1415.

Oliver, J. and R. Babcock. 1992. Aspects ofthe fertilization ecology ofbroadcast

spawning corals: sperm dilution effects and in situ measurements offertilization.

Biological Bulletin 183: 409 - 417.

Oliver, J. K. and B. L. Willis. 1987. Coral-spawn slicks in the Great Barrier Reef:

preliminary observations. Marine Biology 94: 521 - 529.

Oren, D, B. Rickevich and Y. Loya. 1997. Oriented intra-colonial transport of 14C labeled

materials during coral regeneration. Marine Ecology Progress Series 161: 117 ­

122.

Oren, D, Y. Benayahu, H. Lubinevsky and Y. Loya. 2001. Colony integration during

regeneration in the stony coral Faviafavus. Ecology 82: 802 - 813.

Paine, R. T. 1966. Endothermy in bomb calorimetry. Limnology and Oceanography 11:

126 -129.

Paine, R. T. 1971. The measurement and application ofthe calorie to ecological

problems. Annual Review ofEcology and Systematics 2: 145 - 164.

Palumbi, S. R. 2001. The ecology ofmarine protected areas. Pages 509 - 527 in M. D.

Bertness, S. D. Gaines and M. E. Hay, eds. Marine community ecology. Sinauer

Associates Inc., Massachusetts.

Pandolfi, J. M., R. H. Bradbury, E. Sala, T. P. Hughes, K. A. Bjorndal, R. G. Cooke, D.

McArdle, L. McClenachan, M. J. H. Newman, G. Paredes, R. R. Warner and J. B.

C. Jackson. 2003. Global trajectories ofthe long-term decline ofcoral reef

ecosystems. Science 301: 955 - 958.

141 Patton, J. S., J. F. Battey, M. W. Rigler, J. W. Porter, C. C. Black and J. E. Burris. 1983.

A comparison ofthe metabolism ofbicarbonate 14C and acetate 1_ 14C and the

variability ofspecies lipid composition in reefcorals. Marine Biology 75: 121 ­

130.

Pawlik, J. R. 1990. Natural and artificial induction ofmetamorphosis ofPhragmatopoma

lapidosa californica (Polychaeta: Sabellariidae), with a critical look at the effects

ofbioactive compounds on marine invertebrate larvae. Bulletin ofMarine Science

46: 512 - 536.

Pawlik, J. R. 1992. Chemical ecology ofthe settlement ofbenthic marine invertebrates.

Oceanography and Marine Biology Annual Review 30: 273 - 335.

Pearce, C. M and R. E. Scheibling. 1988. Larval settlement in the green sea urchin,

Strongylocentrotus droebachiensis. American Zoologist 28: 71A.

Pearson, R. G. 1981. Recovery and recolonization ofcoral reefs. Marine Ecology

Progress Series 4: 105 - 122.

Pechenik, J. A. 1987. Environmental influences on larval survival and development.

Pages 551- 608 in A. C. Giese, J. S. Pearse and V. B. Pearse, eds. Reproduction

ofmarine invertebrates. Vol 9. Blackwell Scientific, Palo Alto.

Pechenik, J. A. 1990. Delayed metamorphosis by larvae ofbenthic marine invertebrates:

does it occur? Is there a price to pay? Ophelia 32: 63 - 94.

Pechenik, J. A. 1999. On the advantages and disadvantages oflarval stages in benthic

marine invertebrate life cycles. Marine Ecology Progress Series 177: 269 - 297.

Phillipson, J. 1964. A miniature bomb calorimeter for small biological samples. Oikos

15: 1.

142 Pittock, A. B. 1999. Coral reefs and environmental change: adaptation to what? American

Zoologist 39: 10 - 29.

Polacheck, T. 1978. The population biology offour common Hawaiian corals. M.S.

Thesis, Department ofZoology, University ofHawai'i. 151 pp.

Porter, J. W., 1. Muscatine, Z. Dubinsky and P. G. Falkowski. 1984. Primary production

and photoadaptation in light- and shade-adapted colonies ofthe symbiotic coral,

Stylophora pistillata. Proceedings ofthe Royal Society ofLondon, Series B 222:

161 -180.

Pratchett, M. S., N. Gust, G. Goby and S. O. Klanten. 2001. Consumption ofcoral

propagules represents a significant trophic link between corals and reef fish. Coral

Reefs 20: 13 -17.

Raimondi, P. T. and M. J. Keough. 1990. Behavioral variability in marine larvae.

Australian Journal ofEcology 15: 427 - 437.

Raimondi, P. T. and A. N. Morse. 2000. The consequences ofcomplex larval behavior in

a coral. Ecology 81: 3193 - 3211.

Reddingius,1. and P. J. den Boer. 1970. Simulation experiments illustrating stabilization

ofanimal numbers by spreading the risk. Oecologia 5: 240 - 284.

Richmond, R. 1981. Energetic considerations in the dispersal ofPocillopora damicornis

(Linnaeus) planulae. Proceedings ofthe Fourth International Coral Reef

Symposium 2: 153 - 156.

Richmond, R. H. 1985a. Reversible metamorphosis in coral planula larvae. Marine

Ecology Progress Series 22: 181 - 185.

143 Richmond, R. H. 1985b. Variations in the population biology ofPocillopora damicornis

across the Pacific. Proceedings ofthe Fifth International Coral ReefCongress 6:

101 - 106.

Richmond, R. H. 1987a. Energetics, competency, and long-distance dispersal ofplanula

larvae ofthe coral Pocillopora damicornis. Marine Biology 93: 527 - 533.

Richmond, R. H. 1987b. Energetic relationships and biogeographical differences among

fecundity, growth and reproduction in the reefcoral Pocillopora damicornis.

Bulletin ofMarine Science 41: 594 - 604.

Richmond, R. H. 1988. Competency and dispersal potential ofplanula larvae ofa

spawning versus a brooding coral. Proceedings ofthe 6th International Coral Reef

Symposium 2: 827 - 831.

Richmond, R. H. 1990. The effects ofthe El Nino/Southern Oscillation on the dispersal

ofcorals and other marine organisms. Pages 127 - 140 in P. W. Glynn, ed. Global

ecological consequences ofthe 1982-83 EI-Nino-Southern Oscillation. Elsevier

Science Publishers, Amsterdam.

Richmond, R. H. 1993. Coral reefs: present problems and future concerns resulting from

anthropogenic disturbance. American Zoologist 33: 524 - 536.

Richmond, R. H. 1997. Reproduction and recruitment in corals: critical links in the

persistence ofreefs. Pages 175 - 197 in C. Birkeland, ed. Life and death ofcoral

reefs. Chapman & Hall, New York.

Richmond, R. H. and C. L. Hunter. 1990. Reproduction and recruitment ofcorals:

comparisons among the Caribbean, the tropical Pacific, and the Red Sea. Marine

Ecology Progress Series 60: 185 - 203.

144 Rinkevich, B. 1989. The contribution ofphotosynthetic products to coral reproduction.

Marine Biology 101: 259 - 263.

Rinkevich, B. 1995. Restoration strategies for coral reefs damaged by recreational

activities: the use ofsexual and asexual recruits. Restoration Ecology 3: 241 -

251.

Rinkevich, B. 1996. Do reproduction and regeneration in damaged corals compete for

energy allocation? Marine Ecology Progress Series 143: 397 - 302.

Rinkevich, B. and Y. Loya. 1979. The reproduction ofthe Red Sea coral Stylophora

pistillata. I. Gonads and planulae. Marine Ecology Progress Series 1: 133 - 144.

Rinkevich, B. and Y. Loya. 1985. Intraspecific competition in a reefcoral: effects on

growth and reproduction. Oecologia 66: 100 - 105.

Roberts, C. M. 1997. Connectivity and management ofCaribbean coral reefs. Science

278: 1454 - 1457.

Roberts, C. M. 1998. Sources, sinks, and the design ofmarine reserve networks. Fisheries

23: 16 - 19.

Roberts, C. M. 2000. Selecting marine reserve locations: optimality versus opportunism.

Bulletin ofMarine Science 66: 581 - 592.

Rose, M. Rand T. J. Bradley 1998. Evolutionary physiology ofthe cost ofreproduction.

OIKOS 83: 443 - 451.

Rumrill, S. S. and R. A. Cameron. 1983. Effects ofgamma-aminobutyric acid on

settlement oflarvae ofthe black chiton Katharina tunicata. Marine Biology 72:

243 -247.

145 Sakai, K. 1998. Delayed maturation in the colonial coral Goniastrea aspera

(Scleractinia): whole-colony mortality, colony growth and polyp egg production.

Research in Population Ecology 40: 287 - 292.

Sammarco, P. W. 1991. Geographically specific recruitment and postsettlement mortality

as influences on coral communities: the cross-continental shelftransplant

experiment. Limnology and Oceanography 36: 496 - 514.

Sano, M., M. Shimizu and Y. Nose. 1984. Changes in structure ofcoral reeffish

communities by destruction ofhermatypic corals. Pacific Science 38: 51 -79.

Sato, M. 1985. Mortality and growth ofjuvenile coral Pocillopora damicornis

(Linnaeus). Coral Reefs 4: 27 - 33.

Schwartz, J. A, D. A Krupp and V. M. Weis. 1999. Late larval development and onset

ofsymbiosis in the scleractinian coral Fungia scutaria. Biological Bulletin 196:

70 -79.

Sebens, K. P. 1983. Settlement and metamorphosis ofa temperate soft-coral larvae

(Alcyonium siderium Verrill): induction by crustose algae. Biological Bulletin

165: 286 - 302.

Shlesinger, Y. and Y. Loya. 1985. Coral community reproductive patterns: Red Sea

versus the Great Barrier Reef. Science 228: 1333 - 1335.

Smith, S. R. 1992. Patterns ofcoral recruitment and post-settlement mortality on

Bermuda's reefs: comparisons to Caribbean and Pacific reefs. American Zoologist

32: 663 - 673.

146 Smith, S. R. 1997. Patterns ofcoral settlement, recruitment and juvenile mortality with

depth at Conch Reef, Florida. Proceedings ofthe 8th International Coral Reef

Symposium 2: 1197 - 1202.

Smith, S. V. and R. W. Buddemeier. 1992. Global change and coral reefecosystems.

Annual Review ofEcology and Systematics 23: 89 - 118.

Smith, S. V., W. J. Kimmerer, E. A. Laws, R. E. Brock and T. W. Walsh. 1981. Kaneohe

Bay sewage diversion experiment: perspectives on ecosystem responses to

nutritional perturbation. Pacific Science 35: 279 - 395.

Sousa, W. P. 1984. The role ofdisturbance in natural communities. Annual Review of

Ecology and Systematics 15: 353 - 391.

Sousa, W. P. 2001. Natural disturbance and the dynamics ofmarine benthic communities.

Pages 85 - 130 in M. D. Bertness, S. D. Gaines, and M. E. Hay, eds. Marine

community ecology. Sinauer Associates, Inc., Sunderland.

Stehli, F. G. and J. W. Wells. 1971. Diversity and age patterns in hermatypic corals.

Systematic Zoology 20: 115 - 126.

Stimson, J. S. 1987. Location, quantity and rate ofchange in quantity oflipids in tissue of

Hawaiian hermatypic corals. Bulletin ofMarine Science 41: 889 - 904.

Stockhausen, W. T., R. N. Lipcius and B. M. Hickey. 2000. Joint effects oflarval

dispersal, population regulation, marine reserve design, and exploitation on

production and recruitment in the Caribbean spiny lobster. Bulletin ofMarine

Science 66: 957 - 990.

Stoddart, J. A. 1983. Asexual production ofplanulae in the coral Pocillopora damicornis.

Marine Biology 76: 279 - 284.

147 Stoddart, J. A 1986. Biochemical genetics ofPocillopora damicornis in Kaneohe Bay,

Oahu, Hawaii. Pages 133 - 149 in P. L. Jokie1, R. H. Richmond and R. A Rogers,

eds. Coral reefpopulation biology. Hawai'i Institute ofMarine Biology Technical

Report Number 37, SeaGrant Cooperative Report UNIHI-SEAGRANT-CR-86­

Ol.

Strathmann, R. 1974. The spread ofsibling larvae ofsedentary marine invertebrates. The

American Naturalist 108: 29 - 44.

Szmant, A M. 1986. Reproductive ecology ofCaribbean reefcorals. Coral Reefs 5: 43 ­

54.

Szmant, A M. and N. J. Gassman. 1990. The effects ofprolonged "bleaching" on the

tissue biomass and reproduction ofthe reefcoral Montastrea annularis. Coral

Reefs 8: 217 - 224.

Szmant-Froelich, A, M. Reutter and L. Riggs. 1985. Sexual reproduction ofFavia

fragum (Esper): lunar patterns ofgametogenesis, embryogenesis and planulation

in Puerto Rico. Bulletin ofMarine Science 37: 880 - 892.

Szmant-Froelich, AM., P. Yevich and M.E.Q. Pilson. 1980. Gametogenesis and early

development ofthe temperate coral Astrangia danae (Anthozoa: Scleractinia).

Biological Bulletin 158: 257 - 269.

Te, F. T. 1992. Response to higher sediment loads by Pocillopora damicornis planulae.

Coral Reefs 11: 131 - 134.

Thorson, G. 1950. Reproductive and larval ecology ofmarine bottom invertebrates.

Biological Review 25: 1 - 45.

148 Tomascik, T. 1991. Settlement patterns ofCaribbean scleractinian corals on artificial

substrata along a eutrophication gradient, Barbados, West Indies. Marine Ecology

Progress Series 77: 261 - 269.

Tranter, P. R. G., D. N. Nicholson and D. Kinchington. 1982. A description ofspawning

and post-gastrula development ofthe cool temperate coral Caryophyllia smithi.

Journal ofthe Marine Biological Association, United Kingdom 62: 845 - 854.

Unabia, C. R. C. and M. G. Hadfield. 1999. Role ofbacteria in larval settlement and

metamorphosis ofthe polychaetes Hydroides elegans. Marine Biology 133: 55­

64.

Underwood, A. J. and P. G. Fairweather. 1989. Supply-side ecology and benthic marine

assemblages. Trends in Ecology and Evolution 4: 16 - 20.

Underwood, A. J. and M. J. Keough. 2001. Supply-side ecology, the nature and

consequences ofvariations in recruitment ofintertidal organisms. Pages 183 ­

200 in M. D. Bertness, S. D. Gaines and M. E. Hay, eds. Marine community

ecology. Sinauer Associates Inc., Massachusetts.

Van Moorsel, G. W. N. M. 1983. Reproductive strategies in two closely related stony

corals (Agaricia, Scleractinia). Marine Ecology Progress Series 13: 273 - 283.

Van Moorsel, G. W. 1988. Early maximum growth ofstony corals (Scleractinia) after

settlement on artificial substrata on a Caribbean reef. Marine Ecology Progress

Series 50: 127 - 135.

Vance, R. R. 1973. On reproductive strategies in marine benthic invertebrates. The

American Naturalist 107: 339 - 351.

149 Vandenneulen, J. H. 1975. Studies on reefcorals. III. Fine structural changes of

calicoblast cells in Pocillopora damicornis during settlement and calcification.

Marine Biology 31: 69 - 77.

Venneij, G. 1. 2001. Community assembly in the sea, geologic history ofthe living shore

biota. Pages 39 - 60 in M. D. Bertness, S. D. Gaines and M. E. Hay, eds. Marine

community ecology. Sinauer Associates Inc., Massachusetts.

Veron,1. E. N. 1993. A biogeographic database ofhennatypic corals; species ofthe

central Indo-Pacific, genera ofthe world. Australian Institute ofMarine Science

Monograph Series 10. 433 pp.

Veron, J. E. N. 1995. Corals in space and time, the biogeography & evolution ofthe

scleractinia. Comstock/Cornell, Ithaca. 321 pp

Veron, J. E. N. 2000a. Corals ofthe world. Vol. 3. Australian Institute ofMarine Science

and CRR Qld Pty Ltd, Townsville. 490 pp.

Veron, 1. E. N. 2000b. Corals ofthe world. Vol. 1. Australian Institute ofMarine Science

and CRR Qld Pty Ltd, Townsville. 463 pp

Villinski,1. T. 2003. Depth-dependent reproductive characteristics for the Caribbean

reef-building coral Montrastraeafaveolata. Marine Biology 142: 1043 - 1053.

Wallace, C. C. 1983. Visible and invisible coral recruitment. Proceedings ofthe

Inaugural Great Barrier ReefConference, James Cook University Australia: 259­

261.

Wallace, C. C. 1985. Reproduction, recruitment and fragmentation in nine sympatric

species ofthe coral genus Acropora. Marine Biology 88: 217 - 233.

150 Ward, S. 1995a. Two patterns ofenergy allocation for growth, reproduction and lipid

storage in the scleractinian coral Pocillopora damicornis. Coral Reefs 14: 87 - 90.

Ward, S. 1995b. The effect ofdamage on the growth, reproduction and storage oflipids

in the scleractinian coral Pocillopora damicornis (Linnaeus). Journal of

Experimental Marine Biology and Ecology 187: 193 - 206.

Wells, J. W. 1966. Evolutionary development in the scleractinian family Fungiidae.

Symposium ofthe Zoological Society ofLondon 16: 223 - 246.

Wethey, D. S. and J. W. Porter. 1976. Sun and shade differences in productivity ofreef

corals. Nature 262: 281 - 282.

Williams, G. C. 1975. Sex and evolution. Princeton University Press, Princeton, New

Jersey. 200 pp.

Wilson, J. R. and P. L. Harrison. 1997. Sexual reproduction in high latitude coral

communities at the Solitary Islands, eastern Australia. Proceedings ofthe 8th

International Coral Reef Symposium 1: 533 - 538.

Wilson, J. R. and P. L. Harrison. 1998. Settlement-competency periods oflarvae ofthree

species ofscleractinian corals. Marine Biology 131: 339 - 345.

Wyers, S. C. 1985. Sexual reproduction ofthe coral Diploria strigosa (Scleractinia,

Faviidae) in Bermuda: research in progress. Proceedings ofthe Fifth International

Coral Reef Congress 4: 301 - 306.

Zaslow, R. B. and Y. Benayahu. 1996. Longevity, competence and energetic content in

planulae ofthe soft coral Heteroxeniafuscescens. Journal ofExperimental Marine

Biology and Ecology 206: 55 - 68.

151 Zvalinskii, V. 1., V. A. Leletkin, E. A. Titlyanov and M. G. Shaposhnikova. 1980.

Photosynthesis and adaptation ofcorals to irradiance. 2. Oxygen exchange.

Photosynthetica 14: 422 - 430.

152