Parasites and cleaning behaviour in damselfishes Derek Sun

i

Parasites and cleaning behaviour in damselfishes

Derek Sun

BMarSt, Honours I

A thesis submitted for the degree of Doctor of Philosophy at

The University of Queensland in 2015

School of Biological Sciences

ii

Abstract

Pomacentrids (damselfishes) are one of the most common and diverse group of marine fishes found

on coral reefs. However, their digenean fauna and cleaning interactions with the bluestreak cleaner

wrasse, Labroides dimidiatus, are poorly studied. This thesis explores the digenean trematode fauna

in damselfishes from Lizard Island, Great Barrier Reef (GBR), Australia and examines several

aspects of the role of L. dimidiatus in the recruitment of young damselfishes.

My first study aimed to expand our current knowledge of the digenean trematode fauna of

damselfishes by examining this group of fishes from Lizard Island on the northern GBR. In a

comprehensive study of the digenean trematodes of damselfishes, 358 individuals from 32 species

of damselfishes were examined. I found 19 species of digeneans, 54 host/parasite combinations, 18

were new host records, and three were new species (Fellodistomidae n. sp., Gyliauchenidae n. sp.

and Pseudobacciger cheneyae). Combined molecular and morphological analyses show that

Hysterolecitha nahaensis, the single most common trematode, comprises a complex of cryptic

species rather than just one species. This work highlights the importance of using both techniques in

conjunction in order to identify digenean species. The host-specificity of digeneans within this

group of fishes is relatively low. Most of the species possess either euryxenic (infecting multiple

related species) or stenoxenic (infecting a diverse range of hosts) specificity, with only a handful of

species being convincingly oioxenic (only found in one host species). Overall, the trematode

richness in pomacentrid fishes is relatively low compared with that of many other coral reef fish

families. The richness is probably best explained by the small size of most pomacentrids. In terms

of beta diversity, of the total 23 trematode species having now been reported for the GBR in

pomacentrids, just 14 are shared between Heron and Lizard Island. Five species have been recorded

only from Lizard Island and four only from Heron Island. A total of 14 species reported from the

two sites can be classified as core pomacentrid-specific parasites.

I then examined whether the abundance and diversity of damselfish recruits differed on reefs

relative to reefs from which cleaner wrasse had been removed for 12 years. Past studies have shown

that cleaner wrasse enhance abundance and diversity of adult resident damselfishes (species that are

site-attached), however, it is not known whether this effect occurs at recruitment or is driven by

post-settlement events such as migration or differential mortality. I characterised the abundance of

damselfish recruits on reefs with and without cleaner wrasse for five days in each of three lunar

cycles, beginning four days after the new moon (November, December and January). The total

abundance of damselfish recruits and specifically two species, Chrysiptera rollandi and P.

amboinensis, was higher on reefs with cleaner wrasse than on reefs without. However, overall

iii

species diversity of damselfish recruits did not differ. This study shows that the effects of the

presence of cleaner wrasse begin at recruitment.

Although the presence of cleaner wrasse increases the abundance of recruiting damselfish, it

is unclear what mechanism(s) are involved in this process. I conducted aquarium experiments to

examine whether young P. amboinensis preferentially select a microhabitat that is adjacent to a

cleaner wrasse. The study showed that both settlement-stage and juvenile P. amboinensis spend a

greater proportion of time in a microhabitat adjacent to a cleaner wrasse than in one adjacent to a

control wrasse (a non-cleaner wrasse, Halichoeres melanurus) or where no fish was present. More

time was spent next to juvenile than adult cleaner wrasse. I then conducted field observations to

determine if recruits that are in a 1 m radius of a juvenile cleaner wrasse receive any direct benefits

(cleaning) and whether juvenile cleaner wrasse preferentially cleaned larger individuals. The size

and identity of all fish that were cleaned within a 1 m radius of a cleaner were recorded, along with

nearby non-cleaned conspecifics. I showed that only 2 % of the clients that a juvenile cleaner

wrasse clean per 20 min observation were small recruits (< 20 mm, TL) whereas larger

damselfishes comprised 58 % of clients. Juvenile cleaner wrasse also preferred to clean larger

individuals than the median size of the surrounding nearby non-cleaned conspecifics. These results

show that cleaner wrasse serve as a positive cue during microhabitat selection but that direct

benefits to a small recruit are not obvious. The possibilities of indirect rather than direct (cleaning)

benefits are discussed.

I then used sentinel traps to investigate the cumulative number of gnathiid isopods infecting

the ambon damselfish, Pomacentrus amboinensis, on patch reefs with and without cleaner wrasse.

Previous studies may have under surveyed the number of gnathiids on a host related to the presence

or absence of cleaner fish, due to the sampling techniques used (wired cages or handnets). As

gnathiids are temporary parasites, they can drop off a host and return to the benthos before the fish

is sampled; sentinel traps are able to retain such individuals. Nocturnal collections produced almost

three times more gnathiids than diurnal collections. These results are consistent with previous

studies. There was no difference in the number of gnathiids found in sentinel traps placed on reefs

with and without a cleaner wrasse. These results appear initially to be at variance with the many

advantages to fishes reported as being associated with the presence of L. dimidiatus. These

observations lead to the novel hypothesis that the advantage to clients from cleaners is in the

reduction in the length of parasitism rather than the number of infections being reduced.

This research has provided important information regarding the parasites of damselfishes,

from the starting point of the characterisation of the parasite fauna to how cleaning behaviour

affects the ecology of recruiting fish, thus increasing our understanding of recruitment processes on

coral reefs. My study has demonstrated that the richness of damselfishes exceeds that of their

iv

trematode parasites on the GBR although the parasite diversity (11 families) is remarkable. The

work has contributed insight into both host range and species richness, having uncovered both novel

trematode species, and novel hosts for known species. Regarding cleaning interactions, I have

shown that 1) the abundance of damselfish recruits is higher on reefs with cleaner wrasse than on

those without; 2) damselfish recruits preferentially select a microhabitat that is adjacent to a cleaner

wrasse; and 3) that the benefits for young fish to be near cleaners are likely to be indirect rather than

direct. These aspects of my study demonstrates the importance of cleaning mutualism not only for

adult fishes, but for juvenile fishes during the recruitment period.

v

Declaration by author

This thesis is composed of my original work, and contains no material previously published or

written by another person except where due reference has been made in the text. I have clearly

stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical

assistance, survey design, data analysis, significant technical procedures, professional editorial

advice, and any other original research work used or reported in my thesis. The content of my thesis

is the result of work I have carried out since the commencement of my research higher degree

candidature and does not include a substantial part of work that has been submitted to qualify for

the award of any other degree or diploma in any university or other tertiary institution. I have

clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,

subject to the policy and procedures of The University of Queensland, the thesis be made available

for research and study in accordance with the Copyright Act 1968 unless a period of embargo has

been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright

holder(s) of that material. Where appropriate I have obtained copyright permission from the

copyright holder to reproduce material in this thesis.

vi

Publications during candidature

Sun D, Bray RA, Yong R, QY, Cutmore SC, Cribb TH (2014) Pseudobacciger cheneyae n. sp.

(Digenea: Gymnophalloidea) from Weber’s chromis (Chromis weberi Fowler & Bean)

(Perciformes: Pomacentridae) at Lizard Island, Great Barrier Reef, Australia. Systematic

Parasitology 88:141-152

Sun D, Blomberg SP, Cribb TH, McCormick MI, Grutter AS (2012) The effects of parasites on the

early life stages of a damselfish. Coral Reefs 31:1065-1075

Publications included in this thesis

Incorporated as part of Chapter 2: Sun D, Bray RA, Yong R, QY, Cutmore SC, Cribb TH (2014)

Pseudobacciger cheneyae n. sp. (Digenea: Gymnophalloidea) from Weber’s chromis (Chromis

weberi Fowler & Bean) (Perciformes: Pomacentridae) at Lizard Island, Great Barrier Reef,

Australia. Systematic parasitology 88:141-152

TH Cribb, RA Bray and SC Cutmore provided feedback and comments on the manuscript. R. Q-Y

Yong provided editorial advice and assistance with specimen measurements. SC Cutmore assisted

with phylogenetic analyses. D Sun collected and analysed the data and wrote the manuscript.

vii

Contributions by others to the thesis

Chapter 1: TH Cribb provided feedback and comments. D Sun was responsible for the remainder

of the work.

Chapter 2: TH Cribb, RA Bray and R. Q-Y Yong provided feedback and comments on the

manuscript. SC Cutmore assisted with phylogenetic analyses. D Sun collected and analysed the data

and wrote the manuscript.

Part of Chapter 2: TH Cribb, RA Bray and SC Cutmore provided feedback and comments on the

manuscript. R. Q-Y Yong provided editorial advice and assistance with specimen measurements.

SC Cutmore assisted with phylogenetic analyses. D Sun collected and analysed the data and wrote

the manuscript.

Chapter 3: KL Cheney and AS Grutter provided advice on experiment design. KL Cheney, J

Werminghausen, TH Cribb and AS Grutter provided feedback and comments on the manuscript. D

Sun designed and conducted the study, collected and analysed the data and wrote the manuscript.

Chapter 4: KL Cheney and AS Grutter provided advice on experimental design and statistical

procedures. J Werminghausen assisted with data collection. KL Cheney, MI McCormick, MG

Meekan, TH Cribb and AS Grutter provided with feedback and comments on the manuscript. D Sun

designed and conducted the study, collected and analysed the data and wrote the manuscript.

Chapter 5: KL Cheney and AS Grutter provided advice on experimental design and statistical

procedures. J Werminghausen and EC McClure assisted with field data collection. KL Cheney, J.

Werminghausen, EC McClure, MI McCormick, MG Meekan, TH Cribb and AS Grutter provided

editorial and feedback on the manuscript. D Sun designed and conducted the study, collected and

analysed the data and wrote the manuscript.

Chapter 6: TH Cribb provided feedback and comments. D Sun was responsible for the remainder

of the work.

Statement of parts of the thesis submitted to qualify for the award of another degree

None

viii

Acknowledgements

There are many people who need be thanked for their contribution to this project. However, first I

would like to thank my two supervisors, Tom Cribb and Karen Cheney. Tom, your enthusiasm and

knowledge in regards to parasites is remarkable. Throughout the years, including my honours years,

you have guided and supported me and for that I am thankful. Karen, thank you for being

supportive, sharing your knowledge and providing me with invaluable advice. I particularly want to

acknowledge the support from Alexandra Grutter. Your encouragement, passion and enthusiasm for

anything that I did, whether it be small or large made my candidature enjoyable. I’ve learnt so much

and I can’t thank you enough for all of your guidance and advice. Words cannot express my

gratitude to you for everything you have done for me. I would also like to thank both Mark

McCormick and Mark Meekan for reading drafts, providing me with advice and for allowing me to

use their light traps. Also a thank you to both Simon Blomberg and Jeff Hanson for helping me with

all of my questions about statistics.

I am truly grateful for the staff at the Lizard Island Research Station. I thank you for the

support and the friendship that you have all showed me throughout the years. To Lyle, Anne,

Marianne, Lance, Tania and Bob I can’t thank you all enough for helping me with this research, you

have all gone way beyond what is required. Thank you all for the memorable events that we have

all shared. They will never be forgotten.

I am deeply thankful to all the people who have helped me throughout my candidature;

Johanna, Eva, Fabio, Genevieve, Maarten, Russell, Scott, Anne and Andy. I would especially like

to thank Johanna, Eva and Fabio for assisting me in the field, all without ever once complaining for

having to get up at ungodly hours to help me with the light traps or the sentinel traps. Your

assistance is one of the main reasons that I am able to complete this thesis. Genevieve and Maarten

thank you for all of your help in the field, which includes helping me get out of that nally-bin on the

boat. A special thank you goes to Russell, who throughout this PhD has read countless drafts and

has taught me so much about marine parasitology. Scott for helping me with molecular analysis, as

I’m utterly hopeless at it; Matt for teaching me how to put out the light traps and Anne and Andy

for keeping me sane with our regular need for coffee and for your helpful comments and support. I

am also extremely thankful to the members of the Marine Parasitology Lab, Coral Reef Ecology

Lab and the people who occupy room 113 of Goddard throughout my candidature.

I would like to thank the organisations that have provided me with financial support for the

research conducted throughout my candidature. These include the Australian Research Council

Discovery Grant (ARC grant to A. S. G. et al. and K. L. C. and T. H. C. et al.) and Sea World

Research and Rescue Foundation, Australia (SWRRFI) for project funding. I would also like to

ix

thank the International Society for Behavioural Ecology for the student travel award that has

allowed me the opportunity to present this research at the conference in New York.

Finally, I would like to thank my family and friends for always being supportive. Mum and

Dad, thank you for all of your love and financial support. I can’t thank you enough. To my brothers

and sister, William, Henry and Michelle, thank you for always being there for me, even though you

thought that all I did was swim with fishes. I am also grateful to my close friends Karina, Glen,

Justin and Simon for their constant support and kindness. I could always count on you guys to put a

smile on my face. Love all of you guys. Thanks.

x

Keywords

gnathiid isopods, cleaning mutualism, recruitment, trematodes, damselfish, ectoparasites, host-

specificity, Great Barrier Reef, beta diversity

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 060201, Behavioural Ecology, 50%

ANZSRC code: 060205, Marine and Estuarine Ecology, 20%

ANZSRC code: 060301, Animal Systematics and Taxonomy, 30%

Fields of Research (FoR) Classification

FoR code: 0602, Ecology, 70%

FoR code: 0699, Other Biological Sciences, 30%

xi

Table of Contents

Abstract ............................................................................................................................................... ii Declaration by author ........................................................................................................................ v Publications during candidature ..................................................................................................... vi Publications included in this thesis .................................................................................................. vi

Contributions by others to the thesis.............................................................................................. vii Statement of parts of the thesis submitted to qualify for the award of another degree ............ vii Acknowledgements.......................................................................................................................... viii Keywords ............................................................................................................................................ x Australian and New Zealand Standard Research Classifications (ANZSRC) ............................. x Fields of Research (FoR) Classification ........................................................................................... x

Table of Contents .............................................................................................................................. xi List of Figures .................................................................................................................................. xiii

List of Tables ................................................................................................................................... xiv Abbreviations ................................................................................................................................... xv Chapter 1: General Introduction...................................................................................................... 1 1.1 Preamble 2

1.2 Pomacentrids 2

1.3 Parasitism on coral reefs 2

1.4 Parasite diversity on coral reefs 3

1.5 Ecological importance of parasites 5

1.6 Cleaning mutualism 8

1.7 Aims and Thesis structure 11

Chapter 2: Trematodes of northern Great Barrier Reef Pomacentridae: richness, host-

specificity and beta diversity ........................................................................................................... 14 Abstract 15

Introduction 15

Materials and methods 17

Results 19

Discussion 34

Acknowledgments 41



Australia ............................................................................................................................................ 42 Abstract 43

Introduction 43


Gymnophalloidea Odhner, 1905 incertae familiae 47

Discussion 51

Acknowledgements 56

Chapter 3: Presence of cleaner wrasse increases the recruitment of fishes to coral reefs ........ 57 Abstract 58

Introduction 58

Material and methods 59

Results 60

Discussion 64

Acknowledgements 65

xii

Chapter 4: Cleaner wrasse influence habitat selection of young damselfish .............................. 66 Abstract 67

Introduction 68


Results 74

Discussion 78

Acknowledgements 82

Chapter 5: Diel patterns of infection by gnathiid isopods on the ambon damsel, Pomacentrus

amboinensis, and the influence of the presence of cleaner wrasse ............................................... 83 Abstract 84

Introduction 84


Results 93

Discussion 95

Acknowledgements 98

Chapter 6: General Discussion ....................................................................................................... 99 References ....................................................................................................................................... 103 Appendices ...................................................................................................................................... 113

xiii

List of Figures


specificity and beta diversity

Figure 1 Body length and ventral sucker length of the three morphological types of Hysterolecitha

nahaensis 24

Figure 2 Neighbour joining tree based on ITS2 rDNA sequences of Hysterolecitha nahaensis

complex from Lizard Island. Samples are identified by host and dissection number. 25



Australia

Figures 12 Pseudobacciger cheneyae n. sp. ex Chromis weberi from off Lizard Island 49

Figure 3 Relationships between Pseudobacciger cheneyae n. sp. and other Gymnophalloidea,

Bucephalidae and Faustulidae taxa based on Bayesian inference and Maximum Likelihood analyses

of the 28S rDNA dataset. 51

Chapter 3: Presence of cleaner wrasse increases the recruitment of fishes to coral reefs

Figure 1 Total abundance of common damselfishes on experimental reefs. 61

Figure 2 Mean (± SE) abundance of recruiting damselfish species. 62

Figure 3 Diversity of all damselfish recruits per reef. 63

Chapter 4: Cleaner wrasse influence habitat selection of young damselfish

Figure 1 Front view of the laboratory experimental tank design. 71

Figure 2 The percentage of time spent on microhabitat adjacent to fish treatment by damselfish

client Pomacentrus amboinensis. 75

Figure 3 Cleaning interactions of juvenile cleaner wrasse, Labroides dimidiatus and damselfishes.

77

Figure 4 The probability of an individual recruit (< 20 mm) being cleaned by a juvenile cleaner

wrasse, Labroides dimidiatus, in relation to the abundance of a) other fish species and b) abundance

of recruits per 20 min observations. 78


amboinensis, and the influence of the presence of cleaner wrasse

Figure 1 Photograph of the sentinel trap used in the experiment. 91

Figure 2 Individual subcages used to hold Pomacentrus amboinensis in the sentinel traps. 92

xiv

List of Tables



Table 1 Host-parasite associations for pomacentrid fishes and digenean trematodes from Lizard

Island. 32

Table 2 Comparison between data collected from Heron Island (Barker et al. 1994) and Lizard

Island (present study) for digeneans from pomacentrid fishes. 38

Table 3 Core pomacentrid specific digeneans from Heron and Lizard Island. 40



Australia

Table 1 Collection data and GenBank accession numbers for taxa sequenced during this study 46

Table 2 28S rDNA sequence data from GenBank included in phylogenetic analyses 50


Table 1 Ontogenetic stage and size of fish used in each part for laboratory experiment.

Size is mean ± SE, TL. 71


amboinensis, and the influence of the presence of cleaner wrasse

Table 1 Number of traps (with and with no fish) sampled per sampling day, site, reef pair cleaner

presence treatment, and day and night. Total n per host = 382 traps. 90

Table 2 Number of sentinel traps with gnathiids, day and night trap samples and the number of

gnathiids per trap collected from the sentinel trap experiment on Lizard Island 94

xv

Abbreviations

GBR: Great Barrier Reef

PCO: Principle coordinate analysis

TK-HSD: Tukey-Kramer HSD

GLM: General linear model

LMER: Linear mixed-effect models

GLMER: Generalized linear mixed effect model

LRT: Likelihood ratio test

REML: Restricted maximum likelihood

PERMANOVA: Permutational multivariate analysis of variance

NJ: Neighbour joining

PVC: Polyvinyl chloride

SE: Standard error

TL: Total length

ITS2: Internal transcribed spacer 2

rDNA: Ribosomal Deoxyribonucleic acid

mm: Millimetre

h: Hours

s: Seconds

min: Minute

n: Number

m: Metre

1

Chapter 1: General Introduction

2

Chapter 1: General Introduction

1.1 Preamble

This study explores the complex interactions between three key faunal components on the Great

Barrier Reef (GBR): the parasites (both endo- and ectoparasitics) that infect reef fishes, cleaner

wrasse (Labroides dimidiatus) that feed on, and hence remove parasites from fishes, and

pomacentrids fishes (damselfishes), a large and abundant reef fish family. In these interactions,

pomacentrids are contextualised as both hosts to parasites, and clients to the cleaner wrasse. The

fish and their parasites are abundant and diverse; the parasites remain poorly known. The effects of

the parasites are subtle – we never see a pomacentrid dying because of its parasites – but real. The

cleaner wrasse are keystone species whose parasite-eating behaviour has remarkable implications

for the ecosystems overall. Here I review these components and conclude by outlining the studies

that I have completed to further our understanding of them.

1.2 Pomacentrids

The family Pomacentridae (damselfishes) is one of the most abundant and diverse group of marine

fishes that inhabit coral reefs (Randall et al. 1997). Currently, there are 387 species of pomacentrids

recognised globally of which more than 89 are reported for the GBR (Grutter, in review). They

have been placed into five subfamilies: Abudefdufinae, Chrominae, Lepidozyginae (monospecific),

Pomacentrinae and Stegastinae (Cooper et al. 2009). Damselfishes are known to occupy a wide

range of habitats (e.g. rubble, coral and algae substrates) (Medeiros et al. 2010) and collectively

have a rather broad diet, including planktivory, herbivory and omnivory (Meekan et al. 1995).

These traits most likely have contributed to the success of this family on coral reefs. Damselfishes

are thus an important family of fishes that are a major part of and directly influence the structure of

benthic communities (Hixon and Brostoff 1983; Ceccarelli 2007).

1.3 Parasitism on coral reefs

Parasitism is a non-mutualistic symbiotic relationship and is considered one of the most successful

modes of life given the number of parasite species that exist (Poulin and Morand 2000). All groups

of plants and animals are susceptible to parasitism and many groups have adopted parasitism as a

biological strategy. The abundance of parasitism in the marine environment is demonstrated by the

estimate of Rohde (2002) that there are 100,000 species of protistan and metazoan parasites of

marine fishes globally.

3

Coral reefs are the most diverse and complex marine ecosystems and an important

component of the diversity is comprised of parasites. Multiple parasite groups are abundant on coral

reefs including Protista (multiple phyla), Myxozoa, Monogenea, Digenea, Cestoda, Nematoda,

Acanthocephala, Copepoda, and Isopoda (Justine et al. 2010). Justine et al. (2010) predicted that the

1,700 fish species of New Caledonia harbour 17,000 parasite species in total. For trematode worms,

Cribb et al. (2014b) predicted a total of 1,100 ̶ 1,800 species for the GBR and for monogeneans

Whittington (1998) predicted there is at least one species for each GBR fish species. Despite these

high estimates of richness, it is thought, by all estimates, that only a small fraction of species have

been discovered or described (Cribb et al. 2014b). The parasite fauna of many vertebrate groups

(e.g. many fish families) has not been surveyed. In addition, little is known about the ecology or

ecological role of parasites on coral reefs. Relatively few ecological studies have focused on coral

reef parasites relative to the volume of terrestrial and freshwater studies.

1.4 Parasite diversity on coral reefs

Parasitological studies have demonstrated that parasites are highly abundant and diverse in the

organisms that inhabit coral reefs, especially among coral reef fishes (Justine 2010). Here I focus on

one group, the digenean trematodes. These parasites comprise one of the largest groups of internal

metazoan parasites, comprising approximately 18,000 nominal species worldwide (Cribb et al.

2001). So far, 326 species have been described from the GBR, with a total of 1,1001,800 species

predicted for the region (Cribb et al. 2014b), making digeneans a dominant group of internal

parasites of fishes on the GBR. Although studies examining the trematode fauna of GBR fishes

span more than a century, it has only been in the last three decades that the group has been accorded

detailed attention; the present total of 326 described species known now grew from only 30 species

known in 1988 (Cribb et al. 2014b). The majority of known species of trematodes infect the gut,

with a minority specialized in other organs, e.g. Aporocotylidae in the circulatory system (Smith

2002), Gorgoderidae in the urinary bladder (Ho et al. 2014) and Transversotrematidae under the

scales (Hunter and Cribb 2012). They have complex life cycles that require multiple hosts (Poulin

and Cribb 2002). In the most common life cycle form, infection of a molluscan first intermediate

host by a miracidial larva leads to the asexual production of cercariae, which emerge from the

mollusc and infect either the final host directly (Aprorocotylidae and Transversotrematidae), encyst

in the open (e.g. Haploporidae and Gyliauchenidae), or infect a second intermediate host (e.g.

Cryptogonimidae and Bucephalidae). In the second intermediate host the cercaria develops into a

metacercaria, which has the biological goal of being consumed, with its host by the final host. This

mode of infection is called trophic transmission (Brown et al. 2001).

4

The digenean fauna of GBR fishes has been examined systematically, with many of the

major families of fish having been extensively examined. Understanding the host specificity of

digeneans is a critical aspect in understanding transmission in this group. Host specificity among

parasites has been classified into three categories; oioxenicity, where parasites infect a single host

species (e.g. Lepotrema adlardi known only from Abudefduf bengalensis); stenoxenicity, where

species of parasites infect related host species (e.g. Derogenes pearsoni from damselfishes) and

euryxenicity where parasites infect a wide range of hosts (e.g. Aponurus laguncula known from 67

species from 29 fish families) (Miller et al. 2011b).

Most of the survey effort for digeneans of GBR fishes has been concentrated at two

locations, Heron Island on the southern end of the GBR, and Lizard Island in the north, with the

reefs in between little surveyed. The explanation for this is the presence of dedicated research

stations at the two locations, allowing for systematic collection and processing of samples. The first

fish digeneans to be described from the GBR were two species from teleost fishes (Tetraodontidae)

and two from elasmobranchs (Dasyatidae) by Johnston (1913). Of the 326 currently-known species,

just 66 (20 %) species have been reported from both Heron and Lizard Island. However, many fish

families have been examined more thoroughly at one location than the other. An important example

is the Pomacentridae, which have been extensively surveyed at Heron Island, but not at Lizard

Island.

Studies on pomacentrids of the southern GBR (Cribb et al. 1992; Bray et al. 1993b; Bray et

al. 1993a; Barker et al. 1994; Cribb et al. 1994a; Bray and Cribb 1998) demonstrated a fauna of

seven families and 18 species of adult trematodes. These findings were based on a survey at Heron

Island, comprising 312 individuals from 39 species. For Lizard Island on the northern GBR, where

the present studies were based, nine species of trematodes from five families have been reported

(Bray et al. 1993b; Grutter 1994; Cribb et al. 1998; Grutter et al. 2010; Sun et al. 2012), two of

which (Lepotrema oyabitcha and Transversotrema damsella) are not known from Heron Island

(Bray et al. 1993b; Hunter and Cribb 2012). Despite the recording of nine species at Lizard Island,

the fauna there has not been studied systematically and the limited data available suggested that

there are significant differences between the northern and southern GBR. The extent to which the

parasite fauna of the northern and southern GBR is shared by and between fish species is poorly

understood. There has been no systematic study of matched sets of statistically informative samples

of the same species of fishes between the two sites.

5

1.5 Ecological importance of parasites

The importance of parasites to fish health is a complex issue. Although a high proportion of

individual fish everywhere are infected with parasites, pathogenic effects are rarely immediately

evident and are only rarely studied in wild fish. It has been speculated that many parasitic infections

are relatively benign (Candolin and Voigt 2001; Carrassón and Cribb 2014). When closely studied,

however, fish parasites have often been shown to have a variety of negative effects on their host.

They have been shown to affect critical life factors such as host growth rate (Adlard and Lester

1994; Jones and Grutter 2008), body condition (Lemly and Esch 1984), reproduction (Fogelman et

al. 2009; Mensink et al. 2014), population dynamics (Scott and Dobson 1989), and behaviour

(Barber et al. 2000; Roche et al. 2013b).

For coral reef fishes, most parasite related research has been heavily biased towards

diversity studies of the parasites of adult coral reef fishes (Cribb et al. 1994b; Grutter 1994; Morand

et al. 2000) and little is known about the ecological role of most parasites. Few studies have

examined the ecology of internal parasites on coral reef fish (Muñoz et al. 2006; Muñoz et al.

2007). Studies by Grutter et al. (2010) and Sun et al. (2012) showed that there was a difference in

parasite (mostly worms) prevalence with age group in the young damselfishes. These studies also

demonstrated that the presence of internal parasites may have an effect on the growth rate of an

individual at or during settlement. One possible explanation for the limited amount of study of the

effects of internal parasite on the ecology and behaviour of a host is that they are hidden within a

host. There has been more work done on the ecology of ectoparasites, including copepods (e.g.

Finley and Forrester 2003) and especially cymothoids and gnathiids. Adlard and Lester (1994)

showed that barrier reef chromis, Chromis nitida, that were infected with the cymothoid isopod

Anilocra pomacentri had lower growth rate and higher mortality than non-parasitised fish. Mortality

was caused by several factors, including erratic swimming behaviour upon parasitic attachment

leading to increased susceptibility to predation, and gross pathology leading to deterioration in body

condition, particularly in juvenile fishes. Östlund-Nilsson et al. (2005) recorded higher oxygen

consumption, an increase in pectoral fin beat frequency and reduced swimming speed for an

apogonid fish parasitised with Anilocra apogonae. Fogelman and Grutter (2008) showed that three

apogonid species infected by A. apogonae had reduced growth and survival.

Gnathiid isopods are arguably the most ecologically important and certainly the most

carefully studied of the ectoparasites of coral reef fishes. They are abundant in both temperate and

tropical waters (Sikkel et al. 2009) and are common on coral reef fishes (Grutter 1994). They do not

possess a pelagic phase like most other marine invertebrates (Smit and Davies 2004). The life cycle

incorporates three juvenile (larval) stages - three zuphea stages (unfed phases), three praniza stages

(fed phases) and the non-feeding adult (Smit and Davies 2004). These mobile ectoparasites emerge

6

from the benthos as the zuphea phase and locate a host fish. Once engorged on blood and body

fluids, as the praniza phase, they return to the benthos and moult into the next larval phase.

Gnathiids thus exhibit protelian parasitism - parasitic as juveniles and free-living as adults (Smit

and Davies 2004). Unlike other ectoparasites such as copepods and isopods which may be

permanent parasites, gnathiids isopods are temporary parasites and are also often referred to as

micropredators (Lafferty and Kuris 2002).

The mouthparts of gnathiids are highly modified enabling them to attach and feed on the

surfaces of fishes (Monod 1926). Gnathiids attach themselves to the body, fins, eye rims, gill

chambers, nares, lips and the buccal cavity of their fish host (Monod 1926). In feeding on the blood

and tissue fluids from their hosts they can cause either tissue damage or, when present in large

numbers, death (Marino et al. 2004; Hayes et al. 2011). The amount of time required for gnathiids

to become fully engorged varies among both species and developmental stages, with feeding period

being longer for larger species and later juvenile stages (Grutter 2003; Smit et al. 2003; Smit and

Davies 2004). Gnathiids can become engorged in less than 10 minutes (Grutter 2003) but can attach

for up to 10 h (Grutter 2003; Smit et al. 2003; Grutter et al. 2011).

Previous studies on the ecology of gnathiids have demonstrated their importance on coral

reefs. They can have a wide range of effects on adult fish. They can cause tissue damage (Heupel

and Bennett 1999) and gill mucus shedding (Hayes et al. 2011), reduce haematocrit value (Jones

and Grutter 2005), are probable vectors of haemogregarine blood parasites (Davies and Smit 2001;

Davies et al. 2004; Curtis et al. 2013) and viral diseases (Davies et al. 2009), and can even cause

mortality due to blood loss when numbers are high (Mugridge and Stallybrass 1983; Hayes et al.

2011). Infection by gnathiids also influences the behavioural interactions between cleaners and their

clients (Grutter 2001; Sikkel et al. 2004).

In contrast to the relatively detailed knowledge of the role/importance of gnathiids in adult

fish biology, the nature of the interaction between gnathiids and young fish is poorly known. What

is already known about gnathiids on adult fishes (i.e. prevalence, and feeding ecology) (Grutter

1995a,1996a,2003; Grutter 2008) suggests that they may also significantly affect young reef fish.

However, only a handful of studies have investigated their effects on young reef fish. Jones and

Grutter (2008) found that settling Dischistodus perspicillatus damselfish infected with three

gnathiids each night for one week had slower growth, measured by a percent increase in size, than

those infected with one or none. Grutter et al. (2008) recorded mortality (12 % and 16 %) for

settling Neopomacentrus azysron damselfish infected with one or three gnathiids, respectively.

Penfold et al. (2008) found that in the laboratory, gnathiid isopods killed 6.3 % of juvenile

Acanthochromis polyacanthus damselfish that were < 10 mm SL compared with 0 % for uninfected

individuals. Grutter et al. (2011) showed that settling P. amboinensis previously infected with a

7

gnathiid swam more slowly, had higher oxygen consumption, and had reduced survival relative to

uninfected individuals. In the wild, 3.5 % of a population of juvenile P. amboinensis caught at dawn

were infected with a gnathiid and in the laboratory the infection was found to last for up to 10 h

(Grutter et al. 2011). These studies highlight the potential importance of gnathiids in the early life

stages of coral reef fishes and how they may affect successful recruitment to the adult population.

Although numerous studies have examined the emergence of gnathiids in both the

Caribbean and on the GBR in regards to diurnal variation (Jacoby and Greenwood 1988; Grutter

and Hendrikz 1999; Sikkel et al. 2006; Jones and Grutter 2007), little is known as to whether the

presence of cleaner wrasse has a significant effect on the number of gnathiids on reef fish.

Estimating the number of gnathiids on a fish is complicated. As gnathiids are temporary parasites,

they are able to detach themselves from the host when it is disturbed (e.g. host is captured) or can

be eaten by a cleaner fish before the host has been examined for gnathiids. Therefore, studies that

have examined the number of gnathiids in relation to cleaner fish presence may have

underestimated the number of gnathiids on a host. Previous studies that have examined the number

of gnathiids on reef fishes have used sampling methods that may also have overlooked the possible

interaction of cleaner fish and the behaviour of gnathiids. Grutter (1999a) showed that individual

caged thicklip wrasse, Hemigymnus melapterus, had a 4.5-fold reduction in the number of gnathiid

isopods on individuals when on reefs with a cleaner wrasse relative to those on reefs without.

Cheney and Côté (2003) showed that longfin damselfish, Stegastes diencaeus, that were captured

from territories that had access to a cleaning station had lower numbers of gnathiids than

individuals that did not. Although these studies demonstrated that the presence of cleaner fish can

affect the number of gnathiid isopods on reef fish, the findings only provide information in regards

to the gnathiid load on the fish at a given time (time of host captured) and do not take into

consideration the issues raised above in regards to the potential loss of gnathiids.

Sikkel et al. (2011) reported a trap design that is able to quantify the cumulative number of

gnathiids on a host over time. These ‘sentinel’ traps use live ‘bait’ fish to attract gnathiids that are

seeking a host. However, unlike wire cages which have been used in the past (e.g. Grutter 1999a;

Grutter 1999b) from which gnathiids can escape if disturbed or fed, these traps retain gnathiids once

they have entered the trap. The traps also prevent any interaction between cleaner fish and the live

fish inside the traps, thus preventing loss of gnathiids due to removal by cleaner fish. These traps

thus have the capacity to provide an accurate representation of the cumulative number of gnathiids

attracted to a fish over a period of time.

8

1.6 Cleaning mutualism

Positive interspecific interactions are encounters between organisms that occur where at least one

individual benefits while the other is not harmed (Bronstein 1994). Such interactions have been

classified as either mutualism or facilitation/commensalism, with the outcome of the interaction

often creating a more favourable environment, either directly or indirectly, for one or multiple

species (Stachowicz 2001; Bronstein et al. 2004). Mutualistic interactions have been shown to be of

great ecological importance within terrestrial (Bascompte and Jordano 2007), freshwater (Brown et

al. 2012) and marine ecosystems (Knowlton and Rohwer 2003). Some mutualistic relationships can

extend beyond the two species involved in the initial interaction to other species within the

community, and sometimes the overall structure and function of ecosystems (Stachowicz 2001).

Some organisms within mutualistic relationships can have a disproportionate effect on the

success of other species and are known as “keystone” mutualists (Bond 1994). The presence of

keystone mutualists may be pivotal in shaping the structure of communities. Keystone mutualists

are known in both terrestrial (Harms et al. 2000) and aquatic environments (Caley et al. 1996). This

is seen, for example, in the relationship between frugivores (notably mammals and birds) and plants

via seed dispersal, which can lead to enhanced biodiversity of plants (Jordano 2000). On coral reefs,

the bluestreak cleaner wrasse, Labroides dimidiatus, is considered a key stone mutualist because

their presence affects coral reef fish biodiversity (Bshary 2003; Grutter et al. 2003; Waldie et al.

2011) through cleaning symbioses.

Cleaning symbiosis is the removal and subsequent ingestion of ectoparasites, and diseased

and injured tissue by cleaning organisms (Losey 1972; Côté 2000); it occurs in both the terrestrial

and aquatic environment (Poulin and Grutter 1996; Weeks 2000; Cheney and Côté 2001; Becker et

al. 2005; Biani et al. 2009). It has long been held that cleaning symbioses in the marine

environment are mutualistic interactions in which cleaners benefit from the removal of parasites and

tissue from their fish clients (Grutter 1996a; Arnal and Côté 2000; Arnal and Morand 2001) and

clients benefit from a reduced ectoparasite load (Losey 1979; Gorlick et al. 1987; Grutter 1999b). In

the marine environment cleaning behaviour is performed by narrow range of fish and crustaceans

(see list in Poulin and Grutter 1996). Typically cleaners have defined territories, ‘cleaning stations’,

which are visited by organisms, the ‘clients’ (Côté 2000). There is strong evidence to show that

clients do obtain net benefits from the interaction (Bshary et al. 2007; Clague et al. 2011; Ros et al.

2011; Soares et al. 2011). Several studies have shown that the presence of cleaner fish affects the

size and abundance of crustacean parasites on fish (Gorlick et al. 1987; Grutter 1999a; Cheney and

Côté 2001; Grutter and Lester 2002; Clague et al. 2011), increases the size distribution of fish

(Clague et al. 2011; Waldie et al. 2011), and affects the composition of client fish communities

(Bshary 2003; Grutter et al. 2003; Waldie et al. 2011).

9

Labroides dimidiatus is the most common cleaner fish species on the GBR (Randall et al.

1997). At Lizard Island, some individual client fish visit a cleaning station up to 144 times a day

(Grutter 1995a). Individual cleaner wrasses have been shown to consume an average of 1,218 ± 118

parasites, mostly gnathiids, from 2,297 ± 83 fish inspected each day (Grutter 1996a). These studies

suggest that cleaner wrasse presence plays an important role in reducing the number of parasites,

which may ultimately lead to fish choosing to be near a cleaner fish.

On patch reefs around Lizard Island, GBR, the species richness and abundance of visitors

(fish species that move between patch reefs) were reduced when L. dimidiatus individuals were

removed over an 18-month period (Grutter et al. 2003). After 8.5 years of maintaining these same

reefs free of L. dimidiatus, Waldie et al. (2011) found a 23 % and 37 % reduction in species

richness and abundance of all resident fishes (species that are site-attached) relative to reefs that had

cleaners. Furthermore, there was a 66 % reduction in the abundance of juvenile (< 70 mm TL)

visitor fishes on reefs without L. dimidiatus. This reduction in juvenile abundance could explain the

23 % lower abundance and 33 % lower species richness of visitor fishes on these reefs. Although

the benefits of L. dimidiatus are thus well documented for adults and/or visiting reef fishes, its

importance in settlement processes is largely unknown.

The ecological determinants of settlement choice for pelagic coral reef fish larvae are

complex, but dramatically affect population dynamics and the size of future adult populations

(Hoey and McCormick 2004; Heinlein et al. 2010). Selecting an optimum habitat can maximise the

growth and survival of coral reef fish during and after settlement (McCormick and Hoey 2004;

Lecchini et al. 2007a). Factors affecting microhabitat choice include water temperature (Grorud-

Colvert and Sponaugle 2011; Lienart et al. 2014), habitat structure (Feary et al. 2007; Coker et al.

2012b), coral type (Jones 1988; Danilowicz 1996), position on the reef (Srinivasan 2003; Jordan et

al. 2012), and presence of conspecifics (Ohman et al. 1998; Adam 2010). There has been only

limited research on the significance of the presence of heterospecifics in influencing the

microhabitat choice of young reef fish.

A few studies have demonstrated a positive or negative correlation between presence of

heterospecifics and fish settlement. The abundance of some wrasses (Labridae) is highest in the

algal territories of herbivorous damselfishes, possibly due to increased algal cover providing shelter

and food resources at settlement (Green 1994; Green 1998), suggesting correlation rather cause.

Resident piscivorous groupers and moray eels inhibit levels of settlement of damselfish and

acanthurids, but increase the settlement of a labrid, Thalassoma bifasciatum (Almany 2003). As

mortality from predation is often low for cleaners (Côté 2000; Grutter 2004b), it was hypothesised

that juvenile T. bifasciatum may be immune to piscivory due to the species being a cleaner (Almany

2003). In contrast, the presence of two resident damselfishes in the absence of a resident piscivore

10

increases the abundance of acanthurid settlers (Almany 2003). Given the high fish diversity of coral

reefs (Allen 2008), the potential for complex interactions that influence settlement processes is

considerable.

The importance of gnathiids and other external parasites to fish health has also been

demonstrated indirectly by studies such as that of Grutter et al. (2003), which showed a positive

association between cleaner wrasse presence and fish richness and abundance. This relationship

suggested that mobile reef fish actively seek cleaner wrasses in order to have gnathiids and other

parasites removed. Clague et al. (2011) showed that on reefs with cleaner fish, older P. moluccensis

were larger at a certain age than those from reefs with no cleaner fish. They also found that copepod

abundance was higher on fish from reefs with no cleaner fish, but only for larger host fish.

Gnathiids have also been demonstrated to be of importance during settlement, in that even a single

gnathiid infection can decrease the settlement success of P. amboinensis larvae (Grutter et al. 2011).

Thus, if gnathiids are not removed from fish by cleaner fish or gnathiid population densities are

higher on reefs without cleaner fish, there could be a reduction in juvenile fish abundance. It would

thus be an advantage for young reef fish to settle near cleaning stations. However, the role of

cleaning behaviour with respect to influencing the settlement behaviour of juvenile fish remains

relatively unstudied.

As cleaner fish are known to reduce their clients’ ectoparasite loads, proximity to a cleaner

station should be advantageous for resident fish species. Waldie et al. (2011) found a 66 % increase

in juveniles of visiting fish species on reefs associated with a cleaner fish relative to those without

them. However, no study has examined the effect of cleaner wrasse presence on the juveniles of

resident fish species, most of which consist of damselfishes. Previous studies have shown that the

ability to locate suitable habitats during settlement can have significant advantages in terms of

enhancing growth and/or increasing survival of young reef fish (Bonin et al. 2011; Vail and

McCormick 2011; Rankin and Sponaugle 2014). Young reef fish have highly developed visual

(Wenger et al. 2011; Lecchini et al. 2014), olfactory (McCormick et al. 2010; Paris et al. 2013) and

auditory senses (Radford et al. 2011; Huijbers et al. 2012), which are known to play a critical role

when selecting suitable habitats. The humbug damselfish Dascyllus aruanus has been demonstrated

to use olfactory cues released by conspecifics in selecting reefs to settle on (Sweatman 1988). Vail

and McCormick (2011) demonstrated that settling damselfishes use olfaction to recognise predators

and thus avoid settling onto habitats where predators are present.

At present, it is not known whether young reef fish that are found near a cleaning station

have selected these reefs by using visual or other cues. It is known that naïve larvae of Hawaiian

humbug Dascyllus albisella, raised in captivity innately recognise the Hawaiian cleaner wrasse,

Labroides phthirophagus (Losey et al. 1995). That study demonstrated that experience with a

11

cleaner fish is not required for D. albisella to recognise a cleaner fish and to adopt the client ‘pose’.

Although this study showed that young reef fish recognise cleaner fish, no studies have tested

whether cleaner fish presence plays a role in larval habitat choice. A similar mechanism may occur

with young reef fish that are approaching or are making the transition to the benthos, in that they

may partly select reefs based on cues related, directly or indirectly, to cleaner fish presence. As

most coral reef fish (especially damselfishes) are relatively site attached and will not leave their

territory to seek a cleaner fish (Randall et al. 1997), any selection would have to occur while they

are in the process of settling onto the reef or soon after as post-settlement movement (Lewis 1997;

Simpson et al. 2008).

There are other potential benefits besides a reduction in parasite loads for fish that settle in

the vicinity of a cleaner wrasse. Cheney et al. (2008) showed that the presence of a cleaner wrasse

reduces the frequency of aggressive chases by a piscivore towards potential prey species. This

reduction in aggression by predators at cleaning stations may indirectly benefit prey species, by

reducing the risk of being consumed and influencing how long they may remain at a cleaning

station when a predator is present. No doubt such an indirect effect of cleaner fish presence would

benefit such vulnerable young reef fish during settlement, as predation is the major cause of

mortality at that period.

1.7 Aims and thesis structure

The broad aim of this study was to improve understanding of a range of parasite-related issues from

the richness of digenean trematodes to the importance of cleaning behaviour in damselfishes. The

thesis thus explores a series of interrelated topics relating to the parasites of damselfishes and their

ecological implications: (1) the richness and prevalence of digenean trematodes in damselfishes

from the northern GBR, Lizard Island; (2) the effects of cleaner wrasse presence on the abundance

of recruiting damselfishes; (3) the effect of cleaner wrasse presence in a microhabitat choice of a

damselfish; and (4) the cumulative number of gnathiid isopods infecting a species of damselfish

relative to cleaner wrasse presence. These topics are explored through both field and laboratory-

based experiments. The thesis is presented as a collection of chapters written in a style that is

suitable for publication. The thesis consists of six chapters of which four (2 ̶ 5) are in a manuscript

format.

Chapter 2. Trematodes of northern Great Barrier Reef Pomacentridae: richness, host-


12

There are 326 species of trematodes reported so far for the GBR (Cribb et al. 2014b) including 18

species known from damselfishes. Although damselfishes (Pomacentridae) are a major component

of the coral reef fish assemblage, and much is known in regards to their ecology and biology, the

trematode fauna of damselfishes on the northern GBR, has not been examined systematically.

Current knowledge regarding the trematode fauna of damselfishes is mainly from systematic

surveys conducted on Heron Island, southern GBR (Cribb et al. 1992; Bray et al. 1993a; Bray et al.

1993b; Barker et al. 1994). In contrast to damselfishes, other families of fishes have been

extensively surveyed for trematodes at both Heron and Lizard Island - chaetodontids (Bray et al.

1994; McNamara et al. 2012; Diaz et al. 2013), lutjanids (Miller and Cribb 2007; Miller and Cribb

2008) and serranids (Bott and Cribb 2009; Bott et al. 2013). An extensive survey of damselfish

trematodes from Lizard Island, far north GBR formed the first component of my PhD research. This

study led to the description of one new species, Pseudobacciger cheneyae, published as separate

standalone paper (incorporated in chapter 2).

Chapter 3. Presence of cleaner wrasse increases the recruitment of fishes to coral reefs. Although the direct and indirect effects of cleaner fish have been demonstrated for adult visitor reef

fish, they have yet to be examined for resident larval reef fishes. Clague et al. (2011) showed that,

on reefs with cleaner fish, older Pomacentrus moluccensis were larger at a certain age than those

from reefs without cleaner fish. Gnathiids have been demonstrated to be important during

settlement, in that a single gnathiid can decrease the success of settlement of P. amboinensis larvae

significantly (Grutter et al. 2011). Therefore, if gnathiids are not removed from fish by cleaner fish

(direct affect) or gnathiid population densities are higher on reefs without cleaner fish (indirect

affect), there could be a reduction in juvenile fish abundance. Cleaner presence could also affect

larval fish growth or survival through modification of the behaviour of predatory fish and thus the

risk of predation, as there is evidence that predators reduce aggression towards bystanders in the

presence of a cleaner wrasse (Cheney et al. 2008). I examined the effects of cleaner fish presence on

juvenile density using patch reefs that have been experimentally manipulated for 12 years by being

kept free of cleaner wrasses (Clague et al. 2011; Waldie et al. 2011).

Chapter 4. Cleaner wrasse influence habitat selection of young damselfish.

The ability to locate suitable habitats during settlement can confer significant advantages on young

fish by enhancing growth and/or increasing survival (McCormick and Hoey 2004). Young reef fish

have highly developed visual and olfactory senses (Lecchini et al. 2007b) and these play a critical

role in selecting suitable habitats. Naïve captive-raised larval Hawaiian humbug, Dascyllus

albisella, innately recognise Hawaiian cleaner wrasse, Labroides phthirophagus (Losey et al. 1995).

13

Although that study showed that reef fish larvae recognise cleaner fish, no studies have

demonstrated a link between cleaner fish presence and larval microhabitat choice. Parasites such as

gnathiids have already been shown to decrease settlement success of a fish, thus I hypothesise that it

would be an advantage for young reef fish to settle near cleaning stations. I tested whether young

reef fish preferentially choose a microhabitat that is adjacent to a cleaner fish and whether young

fish found at cleaning stations receive direct benefits in the form of cleaning services. If young fish

actively select a microhabitat that is near a cleaner fish, this selection may help increase their

survival after settlement.

Chapter 5. Diel patterns of infection by gnathiid isopods on the ambon damsel, Pomacentrus

amboinensis, and the influence of the presence of cleaner wrasse.

Gnathiid isopods are the most common temporary ectoparasites of coral reef fishes (Grutter and

Poulin 1998), and are the key food of cleaner wrasse (Grutter 1997). The presence of cleaner fish is

known to reduce the number of gnathiids on reef fish (Grutter 1999a; Cheney and Côté 2001) which

could explain the higher abundances of fishes on reefs where cleaner wrasse are present (Grutter et

al. 2003; Waldie et al. 2011) and also the abundance of recruiting fishes. I hypothesise that

predation of gnathiids by cleaner wrasse leads to a reduction in the overall population of gnathiids

on reefs with cleaners relative to those without. Although previous studies have examined the

number of gnathiids on reef fishes (Grutter 1994; Grutter 1999a; Cheney and Côté 2001; Grutter et

al. 2011), these studies only characterise the number of gnathiids on an individual at a specific time

point. As a result, the rate at which gnathiids infect reef fish, and whether that rate changes over the

course of a day, is not understood. Fish might be under surveyed due to gnathiids either dropping

off due to being fully fed or having been eaten by a cleaner fish. In this study I used sentinel traps

that retain gnathiids once they drop off and prevent physical interaction between cleaner wrasse and

client. It is possible that individuals that are vulnerable to gnathiid infection such as young fish are

more commonly infected with gnathiids than has been shown previously. By collecting samples

throughout the day and night (24 hours), I explored the overall cumulative number of gnathiids and

also examined of how it varies with respect to cleaner wrasse presence.

Chapter 6. General Discussion

This chapter integrates the major findings of each study and makes recommendations for future

research.

14

Chapter 2: Trematodes of northern Great Barrier Reef

Pomacentridae: richness, host-specificity and beta

diversity

15

Chapter 2: Trematodes of northern Great Barrier Reef Pomacentridae:

richness, host-specificity and beta diversity

Derek Sun, Rodney A. Bray, Russell Q-Y. Yong, Scott C. Cutmore & Thomas H. Cribb

Abstract

Species richness and host specificity of the digenean fauna of the family Pomacentridae was studied

by examining 358 individuals of 32 species of pomacentrids from Lizard Island, northern Great

Barrier Reef (GBR). In total, 19 clear morphotype species of digeneans, 54 host/parasite

combinations, 19 new host distributions, 18 new host records and three new species,

Fellodistomidae n. sp., Gyliauchenidae n. sp., and Pseudobacciger cheneyae Sun, Bray, Yong,

Cutmore & Cribb, 2014, were recorded from this large and important group of coral reef fishes.

Combined morphological and molecular analysis of specimens initially identified as Hysterolecitha

nahaensis suggested that on the basis of morphology it is possible to recognise three species and

from molecular analysis as many as seven. The two sources of data could not be completely

reconciled and the system requires further work. For the purpose of this analysis all such forms are

identified as “H. nahaensis complex”. Further examination of the host-parasite relationship showed

that host specificity is relatively low; only three species are apparently oioxenic whereas the

majority possessed either stenoxenic or euryxenic host specificity. Overall, a total of 23 species of

digenean trematodes have now been reported in pomacentrids from the GBR of which 13 are

present at both Lizard and Heron Island. Five and four species have been reported only from Lizard

and Heron Island, respectively. Lastly, 14 of the 23 species have been reported from only

pomacentrids. The richness that was recorded is similar to that reported previously at Heron Island.

This study suggests that the diversity of pomacentrids species far exceeds that of their digenean

parasites at Lizard Island.

Introduction

Damselfishes (Pomacentridae) are one of the most diverse families of marine fishes occurring on

coral reefs (Randall et al. 1997). Currently, there are 387 species of pomacentrids recognised

globally in both marine and brackish systems (Froese and Pauly 2012). The family is divided into

16

five subfamilies: Abudefdufinae, Chrominae, Lepidozyginae (monospecific), Pomacentrinae and

Stegastinae (Cooper et al. 2009); all five subfamilies are represented on the GBR and more than 80

species are known in the region (Randall et al. 1997).

Habitat selection and a wide diversity of diet preferences of damselfishes may have aided

diversification of this family on coral reefs. The large diversity within this group has allowed

damselfishes to occupy and exploit a range of microhabitats such as living corals (Wilson et al.

2008), rubble/dead coral (Bay et al. 2001), and algae-dominated areas (Medeiros et al. 2010), as

well as exploiting both benthic and pelagic niches (Randall et al. 1997). Damselfishes can be

divided into three major trophic guilds: planktivores, herbivores and omnivores (Cooper et al.

2009); the scope for exploitation by trophically transmitted parasites, including trematodes, in this

family is thus substantial.

Host specificity is an important aspect of the biology of parasites (Adamson and Caira

1994). Host specificity of parasites can be classified into three broad categories: oioxenicity, where

an individual parasite species is known to only infect a single host species; stenoxenicity, where the

parasite infects multiple phylogenetically related species; and euryxenicity, where a species infects

a diverse range of hosts united by ecology and physiology rather than taxonomy (Miller et al.

2011b). Miller et al. (2011b) showed that 60 % of the trematodes then known from the GBR were

oioxenic, 32 % were stenoxenic and just 8 % were euryxenic. For the Pomacentridae specifically,

Barker et al. (1994) found that, of the 18 species of digeneans known from pomacentrids at Heron

Island, six [Bivesicula unexpecta Cribb, Bray & Barker, 1994, Haplosplanchnidae sp., Lepotrema

adlardi (Bray, Cribb & Barker, 1993), Lepotrema sp. 1, Lepotrema sp. 2 and Steganoderma gibsoni

Cribb, Bray &Barker, 1992] were oioxenic, three (Derogenes pearsoni Bray, Cribb & Barker, 1993,

D. pharyngicola Bray, Cribb & Barker, 1993 and Hysterolecitha heronensis Bray, Cribb & Barker,

1993) were strictly stenoxenic, another two [Hysterolecitha nahaensis Yamaguti, 1942 and

Schikhobalotrema pomacentri (Manter, 1937)] were predominantly stenoxenic (i.e. primarily found

in one group of closely-related hosts, but occasionally found in hosts from unrelated families) and

seven [Aponurus laguncula Loos, 1907, Lepotrema clavatum (Ozaki, 1932), Lecithaster stellatus

Looss, 1907, Macvicaria heronensis Bray & Cribb, 1989, Preptetos cannoni Barker, Bray & Cribb,

1993, P. xesuri (Yamaguti, 1940) Pritchard, 1960 and Thulinia microrchis (Yamaguti, 1934) Bray,

Cribb & Barker, 1993] were euryxenic. The host range thus varied greatly between species, with

one, Hysterolecitha nahaensis, found in 24 pomacentrid species.

Although damselfishes have been the focus of many ecological studies in reef fishes their

parasitic fauna remains incompletely known. The first dedicated investigation into damselfish

digenean fauna was conducted at Heron Island in the early 1990s. A total of 39 pomacentrid species

was examined, resulting in the discovery of 18 species of trematodes and the description of 8 new

17

species (Cribb et al. 1992; Bray et al. 1993b; Bray et al. 1993a; Bray and Cribb 1998). Relatively

few studies have been conducted elsewhere on the GBR; of these, work was primarily conducted at

Lizard Island. However, unlike at Heron Island, these studies focused on individual species of

damselfishes or a particular family of digenean trematode (Grutter et al. 2010; Hunter and Cribb

2010; Sun et al. 2012; Sun et al. 2014). Thus, the trematode fauna of damselfishes at other locations

on the GBR, including Lizard Island, can be considered to be under-studied. Studies conducted

elsewhere in the world (e.g. Yeo and Spieler 1980; Dyer et al. 1985) have demonstrated a relatively

depauperate digenean fauna in damselfishes.

This is the first study to have systematically examined the diversity and host specificity of

digenean trematodes in the family Pomacentridae from Lizard Island in the northern Great Barrier

Reef and compared its nature relative to the existing known fauna at Heron Island reef 1,184 km to

the south.

Materials and methods

Specimen collection

Between 2011 and 2012, pomacentrids were collected from Lizard Island (1440S, 14528E),

Great Barrier Reef, Australia. These fish were collected using a barrier net, spear gun or clove oil,

and were returned to the research station and euthanized prior to dissection via cranial pithing. As

transversotrematids live beneath the scales of fish the body was soaked in vertebrate saline for a

minimum of 10 minutes. This allowed any transversotrematids to fall to the bottom of the container.

The heart was removed from the body and examined carefully for aporocotylids. The contents of

the gut cavity were removed and examined under a dissecting microscope using fine scissors and

forceps. A “gut wash” was conducted in an attempt to dislodge parasites attached to the gut wall

(Cribb and Bray 2010). Specimens were killed and heat-fixed by pipetting the worms into 0.85 %

saline just off the boil and transferred to Eppendorf tubes containing 70 % absolute ethanol (Cribb

and Bray 2010).

Morphological analysis

Preserved worms were stained using Mayer’s haematoxylin, destained in 1 % hydrochloric acid,

neutralised with 1 % ammonia solution, dehydrated through a graded series of ethanols (50, 70, 80,

90 and 100 %), cleared in methyl salicylate and then mounted on slides in Canada balsam. All

measurements were made using a digital camera and the programme SPOT Advanced (version 4.6).

Voucher specimens are lodged within the Queensland Museum.

18

Molecular analysis

All voucher specimens are paragenophores (sensu Pleijel et al. 2008). All vouchers are

lodged in the Queensland Museum, Brisbane, Australia. Total genomic DNA was extracted using

standard phenol/chloroform extraction techniques (Sambrook and Russell 2001). The D1-D3

regions of 28S nuclear ribosomal DNA was amplified using the primers LSU5 (5'-TAG GTC GAC

CCG CTG AAY TTA AGC-3') (Littlewood et al. 2000) and ECD2 (5'-CTT GGT CCG TGT TTC

AAG ACG GG-3') (Littlewood et al. 2000) and the ITS2 region using the primers 3S (5'-GGT ACC

GGT GGA TCA CGT GGC TAG TG-3') (Bowles et al. 1993) and ITS2.2 (5'-CCT GGT TAG TTT

CTT TTC CTC CGC-3' ) (Cribb et al. 1998) PCR for both the 28S and ITS2 regions was performed

with a total volume of 20 μl consisting of approximately 10 ng of DNA, 5μl of 5x MyTaq Reaction

Buffer (Bioline), 0.75 μl of each primer (10 pmols) and 0.25 μl of Taq DNA polymerase

(BiolineMyTaq™ DNA Polymerase), made up to 20 μl with Invitrogen™ ultraPURE™ distilled

water. Amplification was carried out on a MJ Research PTC-150 thermocycler. The following

profile was used to amplify the 28S region: an initial 95°C denaturation for 4 min, followed by 30

cycles of 95°C denaturation for 1 min, 56°C annealing for 1 min, 72°C extension for 2 min,

followed by a single cycle of 95°C denaturation for 1 min, 55°C annealing for 45 sec and a final

72°C extension for 4 min. The following profile was used to amplify the ITS2 region: an initial

single cycle of 95°C denaturation for 3 min, 45°C annealing for 2 min, 72°C extension for 90 sec,

followed by 4 cycles of 95°C denaturation for 45 sec, 50°C annealing for 45 sec, 72°C extension for

90 sec, followed by 30 cycles of 95°C denaturation for 20 sec, 52°C annealing for 20 sec, 72°C

extension for 90 sec, followed by a final 72°C extension for 5 min. Amplified DNA was purified

using a BiolineISOLATE II PCR and Gel Kit, according to the manufacturer’s protocol. Cycle

sequencing of purified DNA was carried out for the ITS2 rDNA region using the forward primer

GA1 (5'-AGA ACA TCG ACA TCT TGA AC-3') (Anderson and Barker 1998) and the reverse

primer ITS2.2 and for the 28S rDNA region using the same primers used for PCR amplification as

well as the additional primer 300F (5'-CAA GTA CCG TGA GGG AAA GTT-3')(Littlewood et al.

2000). Cycle sequencing was carried out at the Australian Genome Research Facility using an

AB3730xl capillary sequencer. Sequencher™ version 4.5 (GeneCodes Corp.) was used to assemble

and edit contiguous sequences. Sequences were aligned using Muscle implemented with MEGA

5.2. Differences between sequences were explored by creating distance matrices and neighbour

joining (NJ) trees.

19

Results

A total of 358 individual fish were examined from 32 species of pomacentrids from Lizard Island.

In total, 19 clear morphotype species of digeneans, 54 host/parasite combinations, 18 new host

records, 19 new host parasite/locality records and three new species were recorded (see Table 1).

Below, the details of observations relating to the 19 species are summarised.

Bivesiculidae Yamaguti, 1934

Bivesicula unexpecta Cribb, Bray & Barker, 1994

Host: Acanthochromis polyacanthus (Bleeker, 1855).

Locality: Lizard Island

Voucher specimens: QM XXXXX

Remarks: This species was described from Acanthochromis polyacanthus by Cribb et al. (1994a) at

Heron Island. Sequence data for this species from the same host species was subsequently reported

for both Heron and Lizard Islands (Cribb et al. 1998). The present specimens are entirely consistent

with the original description of this species. Bivesicula unexpecta remains the only bivesiculid

species to have been reported as an adult from a pomacentrid and A. polyacanthus remains the only

known host species. Immature bivesiculids were found in several other pomacentrids in this study,

but these were all consistent with Bivesicula claviformis Yamaguti, 1934, a species that matures in

serranid fishes and has immature stages in the intestine of a range of fishes preyed on by serranids

(Cribb et al. 1998).

Derogenidae Nicoll, 1910

Derogenes pearsoni Bray, Cribb & Barker, 1993

Hosts: Chrysiptera cyanea (Quoy & Gaimard, 1825), Pomacentrus adelus Allen, 1991, Stegastes

apicalis (De Vis, 1885).



Remarks: Our specimens of this species match those reported by Bray et al. (1993a) from the

damselfishes Amblyglyphidodon curacao (Bloch, 1787), Amphiprion akindynos Allen, 1972,

Plectroglyphidodon lacrymatus (Quoy & Gaimard, 1985), Pomacentrus chrysurus Cuvier, 1830, P.

moluccensis Bleeker, 1853, P. tripunctatus Cuvier, 1830, and an unidentified species of

Pomacentrus sp. from Heron Island, in the southern GBR. Sun et al. (2012) reported D. pearsoni

from Pomacentrus amboinensis Bleeker, 1868 from Lizard Island. The three hosts reported here

represent new host records. Derogenes pearsoni differs from the other species of Derogenes found

in GBR pomacentrids, D. pharyngicola Bray, Cribb & Barker, 1993, in having shorter caeca, the

20

structure of the sinus-sac, the location of the posterior genital pore and the development of the

uterus in the forebody (Bray et al. 1993a). Derogenes pharyngicola was not detected in the present

study.

Faustulidae Poche, 1926

Pseudobacciger cheneyae Sun, Bray, Yong, Cutmore & Cribb, 2014

Host: Chromis weberi Fowler & Bean, 1928.


Voucher specimens: QM G 234368G 234387

Remarks: This species was described from Chromis weberi by Sun et al. (2014) (as part of this

study, published paper appended to this chapter) from Lizard Island. Pseudobacciger cheneyae is

the first faustulid to have been reported from a pomacentrid, with other members of the genus

known primarily from clupeids, but also from pomacanthids and sparids (Parukhin 1976b; Machida

1984; Dimitrov et al. 1999).

The systematics of this species, its family and higher classification are problematic. Although the

genus Pseudobacciger is classified within the family Faustulidae, phylogenetic analyses show that

P. cheneyae is more closely related to members of the superfamily Gymnophalloidea, which

includes, amongst others, the Fellodistomidae and Tandanicolidae, to which P. cheneyae is sister

(Sun et al. 2014). Currently, P. cheneyae displays oioxenic host-specificity in that it has only been

recorded from one host species, despite the ecology and behaviour of C. weberi resembling that of

other planktivorous damselfishes.

Fellodistomidae Nicoll, 1909

Fellodistomidae n. sp.

Host: Chrysiptera rollandi (Whitley, 1961).



Remarks: This species was represented by just a single specimen from 1 of 9 Chrysiptera rollandi

examined. The specimen agrees with the family Fellodistomidae in its smooth tegument, restricted

vitellarian and Y-shaped excretory vesicle but not with any presently recognised genus. The fact

that only a single specimen was found casts doubt on its status as a genuine parasite of

pomacentrids, but no such form has been seen in any of thousands of other fishes examined at

Lizard Island. The status of this form remains unclear.

21

Gyliauchenidae Fukui, 1929

Gyliauchenidae n. sp.

Host: Stegastes apicalis (De Vis, 1885).


Deposited specimens: QM XXXXX

Remarks: The Gyliauchenidae is a small family of trematodes in the superfamily Lepocreadioidea,

whose members primarily infect herbivorous reef fishes (Bray et al. 2009). Only one species,

Gyliauchen pomacentri Nahhas & Wetzel, 1995, has been previously reported from a pomacentrid

fish, Pomacentrus philippinus from off Suva, Fiji (Nahhas and Wetzel 1995). A description of this

new species is being prepared in collaboration with Dr Kathryn Hall of the Queensland Museum.

Haplosplanchnidae Poche, 1926

Schikhobalotrema pomacentri Skrjabin & Guschanskaja, 1955

Hosts: Amblyglyphidodon curacao (Bloch, 1787), Chromis viridis (Cuvier, 1830), Pomacentrus

moluccensis Bleeker, 1853.

Voucher specimen: QM XXXXX

Remarks: The family Haplosplanchnidae is known from a wide range of marine teleosts, with three

species from the genus Schikhobalotrema Skrjabin & Guschanskaja, 1955, reported from

pomacentrids. Schikhobalotrema crassum Pritchard & Manter, 1961, was described from Stegastes

fasciolatus (Ogilby, 1889) from Hawaii, S. robustum Pritchard & Manter, 1961 was described from

S. fasciolatus as well as two acanthurid and one chaetodontid species, also from Hawaii (Pritchard

and Manter 1961), and S. pomacentri (Manter, 1937) Skrjabin & Guschanskaja, 1955, has been

reported from 17 species of pomacentrid (plus one labrid species), from both Pacific and Atlantic

coasts of North, Central and South America, as well as Heron Island (Cribb et al. 1994). The

present specimens were consistent with S. pomacentri. The taxonomy of this genus has not been

considered in any detail for specimens from the GBR and we think it is likely that parallel

molecular studies will be needed to resolve it. The identification of S. pomacentri should be

considered provisional.

Lecithasteridae Odhner, 1905

Aponurus laguncula Looss, 1907

Host: Abudefduf whitleyi Allen & Robertson, 1974



22

Remarks: This species is reported as being cosmopolitan, being found in various locations

worldwide and infecting over 60 different host species from 28 families (Bray & Mackenzie, 1990).

Bray et al. (1993a) first reported and described this species from the Pomacentridae, having

discovered it in Pomacentrus moluccensis from Heron Island, while Grutter et al. (2010) and Sun et

al. (2012) reported it from P. moluccensis and P. amboinensis from Lizard Island, respectively. Our

discovery of A. laguncula from the Whitley’s sergeant, Abudefduf whitleyi, represents a new host

record for this species. This species differs from the closely-related A. rhinoplagusiae Yamaguti,

1934, which also infects pomacentrids, in having a smaller seminal receptacle (Bray et al. 1993a).

Hysterolecitha heronensis Bray, Cribb & Barker, 1993

Host: Pomacentrus amboinensis Bleeker, 1868



Remarks: Our specimens of this species match the original description by Bray et al. (1993a) from

five species of Pomacentridae from Heron Island, Pomacentrus amboinensis, P. moluccensis, P.

nigromarginatus Allen, 1973, P. philippinus Evermann & Searle, 1907, and an unidentified

Pomacentrus sp. It was later reported from P. amboinensis from Lizard Island (Sun et al. 2012).

Hysterolecitha heronensis is similar to H. nahaensis Yamaguti, 1942, with which it may have been

confused historically (Bray et al. 1993a). It is distinguished from H. nahaensis by possessing

notably digitiform vitelline lobes, a larger sinus-sac, a greater average sucker ratio and more distinct

genital atrium. It is noticeable that at Lizard Island this species has been found only in species of

one pomacentrid genus, Pomacentrus.

Hysterolecitha nahaensis Yamaguti, 1942 complex

Host: Abudefduf bengalensis (Bloch, 1787), Abudefduf sexfasciatus (Lacépède, 1801),

Acanthochromis polyacanthus (Bleeker, 1855), Amblyglyphidodon curacao (Bloch, 1787), Chromis

atripectoralis Welander & Schultz, 1951, Dascyllus reticulatus (Linnaeus, 1758), Dischistodus

melanotus (Bleeker, 1858), Neopomacentrus melas (Cuvier, 1830), Pomacentrus adelus Allen,

1991, Pomacentrus amboinensis Bleeker 1868, Pomacentrus chrysurus Cuvier, 1830, Pomacentrus

moluccensis Bleeker, 1853, Pomacentrus nagasakiensis Tanaka, 1917, Pomacentrus pavo (Bloch,

1787), Pomacentrus wardi Whitley, 1927.



Remarks: Specimens recorded here as “Hysterolecitha nahaensis complex” created a special

identification problem. This species was described by Yamaguti (1942) from two species of

23

Pomacentridae and one of Scaridae with Dascyllus aruanus (Pomacentridae) as the type-host. From

the same general region it was subsequently re-reported from Japanese waters (Ichihara 1974; Dyer

et al. 1988), south Vietnam (King 1964), Indonesia (Sulawesi) (Yamaguti 1953), and the Great

Barrier Reef (Lester and Sewell 1990; Bray et al. 1993a; Barker et al. 1994; Sun et al. 2012). It has

also been reported from “Southern Seas” by Parukhin (1989) and from the Mediterranean by Rizk

et al. (1996). In Asia and waters from the Great Barrier Reef, records of this species have been

overwhelmingly from pomacentrids although there are single records from an acanthurid, a lobotid

and a scarid. Bray et al. (1993a) reported 23 species of Pomacentridae (and no other families of

fishes) as hosts for this species on the GBR. In that study a multivariate analysis of 62 specimens

identified some variation in which specimens from Dascyllus reticulatus showed the greatest

tendency to clumping, but the variation was considered inconclusive and all the specimens were

identified as a single species. Bray et al. (1993a) considered the record of Parukhin (1989) from

“Southern Seas” to be unconvincing.

In the present study H. nahaensis was by far the most common infection. However,

morphological examination of 105 specimens initially identified as H. nahaensis showed evidence

of representing three clearly distinct species as follows:

Type 1. Multiple specimens from Dascyllus reticulatus, two from Pomacentrus moluccensis, and

single specimens from Acanthochromis polyacanthus and Stegastes apicalis were distinguishable

from all other samples on the basis of their ventral suckers being proportionally distinctly smaller

than those of comparable-sized individuals from all other fishes (Figure 1).

Type 2. Eight morphological specimens (1 ex N. melas, 3 ex P. amboinensis, 2 ex P. moluccensis

and 2 ex P. wardi) differed from all other specimens in the sample in the ventral sucker being

proportionally far larger than those of comparable-sized specimens (Figure 1). The sucker length

ratio in these specimens ranged from 4.31-5.90. In addition, in these specimens the testes are

unusually close to the ventral sucker, the vitellarium is proportionally large, and the oral sucker is

proportionally small.

Type 3. This was by far the most common type (71 measured specimens) and it was found in 14

pomacentrid species Abudefduf sexfasciatus, Acanthochromis polyacanthus, Amblyglyphidodon

curacao, Dascyllus reticulatus, Dischistodus melanotus, Neoglyphidodon melas, Pomacentrus

adelus, P. amboinensis, P. chrysurus, P. moluccensis, P. nagasakiensis, P. pavo and P. wardi. The

size of the ventral sucker of these specimens was clearly intermediate between those of

24

Morphotypes 1 and 2 (Figure 1). Examination of specimens and comparison of basic measurements

suggested no consistent basis for further subdivision of this form.

Figure 1 Body length and ventral sucker length of the three morphological types of Hysterolecitha

nahaensis.

In addition to the morphological examination of specimens, ITS2 rDNA sequences were generated

for 18 specimens from 7 pomacentrid species. These formed seven distinct genotypes (Genotypes

A-G) which each differed from each other by at least 7 bp. Three of the genotypes were represented

by a single sequence; the others had 2, 2, 5 and 6 replicates. In numerous studies of trematodes of

marine fishes consistent differences in ITS2 rDNA sequences at the level of 7 bp have been found

consistent with the recognition of separate species (e.g. McNamara et al. 2014). In the replicated

sequences there was no variation except for a single bp difference in one specimen of the genotype

with 5 replicates. Distinctions between these taxa are illustrated in an NJ tree (Figure 2). We are

able to make the following observations about these genotypes:

25

Figure 2 Neighbour joining tree based on ITS2 rDNA sequences of Hysterolecitha nahaensis

complex from Lizard Island. Samples are identified by host and dissection numbers.

Genotype A. 5 sequences ex Dascyllus reticulatus and 1 ex Stegastes apicalis. This genotype

corresponds with Morphotype 1 in that all sequences from D. reticulatus were of this genotype and

the great majority of morphological specimens were of Morphotype 1. The single sequence and

morphological specimen from S. apicalis corresponded in the same way.

Genotype B. Two identical sequences ex Chromis atripectoralis. No complete morphological

specimens corresponding to this genotype are available; partial paragenophore specimens are

consistent with H. nahaensis but are unusually large. One paragenophore specimen with part of the

posterior body removed measures 1,816 µm to the vitellarium, far longer than any other specimen

in the complete collection for which the maximum measurement is 1,505 µm. The combination of

genetic and morphological distinction are presumably consistent with the presence of a distinct

species.

Genotype C. A single completely distinct sequence ex Pomacentrus moluccensis. This is one of

three genotypes reported from this species; it is the repeated sharing of hosts by multiple genotypes

26

that makes delineation in this group especially problematic. The single morphological specimen

from the dissection corresponding to the sequenced specimen is a standard and unremarkable

Morphotype 3. It is not presently possible to relate this genotype to a distinct morphotype.

Genotype D. A single completely distinct sequence ex Pomacentrus adelus. This host is also

represented in Genotype E. There is no corresponding morphological specimen from the fish from

which the sequenced specimen was obtained. It is not presently possible to relate this genotype to a

distinct morphotype.

Genotype E. Five sequences from Chromis viridis, Amblyglyphidodon curacao, Pomacentrus

adelus and P. moluccensis (4 x identical, the C. viridis sample differs at 1 bp). The great bulk of the

specimens from these hosts relate to Morphotype 3 including two sequenced from the same

dissections as are represented among the morphological material. This combination presumably

reflects a distinct species and Genotype C and Morphotype 3 are here provisionally identified as the

same species. However, this identification is compromised by the fact that other genotypes have

been found in two of the fish species (P. adelus and P. moluccensis).

Genotype F. A single completely distinct sequence ex Abudefduf whitleyi. No morphological

specimens at all are available from this host. The few specimens from Abudefduf sexfasciatus (for

which there are no sequences) are slightly distinctive in having unusually large ventral suckers for

the specimens in Morphotype 3. It is not presently possible to relate this genotype to a distinct

morphotype, but it seems possible that it represents a distinct “Abudefduf” species.

Genotype G. Two identical sequences ex P. moluccensis from two separate individual fish. One of

those fish had infections of both Morphotype 2 and 3, the other only Morphotype 3 among the

examined morphological specimens. It is not presently possible to relate this genotype to a distinct

morphotype.

There are no ITS2 rDNA sequences for specimens of Hysterolecitha nahaensis available for

comparison from any other locality; indeed, there are none for the genus except for an 18S rDNA

sequence of H. nahaensis (see Blair et al. 1998). Perusal of specimens identified as H. nahaensis by

Bray et al. (1993a) from pomacentrids from Heron Island suggests that there is almost certainly

more than one species represented.

We conclude that it is impossible to be definitive about the richness of Hysterolecitha in

GBR pomacentrids. The subject requires further intensive study, especially the sequencing of more

paragenophore specimens. It is noteworthy that the type-host of H. nahaensis, Dascyllus aruanus,

27

was surveyed thoroughly (n = 20) but had no infections of any sort at Lizard Island. For the

purposes of the present analysis we have chosen to record all the forms discussed above as “H.

nahaensis complex”. This is for two reasons. First, although some of the putative taxa within the

complex appear to be clearly distinct, there are many grey areas in which the differences between

putative taxa are completely unresolved; we are unable to propose a convincing hypothesis for the

number of species present although we conclude that there are at least four, but potentially as many

as seven. Given that three genotypes were represented by only single sequences and that specimens

from many pomacentrid species have not been sequenced at all, it seems possible that even more

diversity remains undetected in this system. Second, a key aim of this study was to compare the

trematode fauna of pomacentrids of the northern GBR with that reported previously for the southern

GBR. We conclude that the identifications of the H. nahaensis complex for the southern GBR are

also confounded by the presence of a complex of species and reiterate that no molecular data at all

are available from there. We thus propose to identify all these forms as “H. nahaensis complex”

with a view to future resolution; determining the extent to which the fauna of the northern and

southern GBR is comparable for this complex is thus beyond the scope of the present study.

Lecithaster stellatus Looss, 1906

Hosts: Chrysiptera rollandi (Whitley, 1961), Pomacentrus amboinensis Bleeker, 1868,

Pomacentrus moluccensis Bleeker, 1853.



Remarks: Our specimens of this species match the description by Bray et al. (1993a). This species

has a large distribution range, having been recorded throughout the Indo-west and west-central

Pacific, as well as from the Mediterranean Basin, in both tropical and temperate waters (Bray et al.

1993a). It has been recorded from 17 fish families so far (reviewed by Bray et al. 1993a), including

at least one record from fresh water in anadromous fishes (Salmonidae) (Bykhovskaya-Pavlovskaya

et al. 1962). Bray et al. (1993a) reported L. stellatus from 11 species of Pomacentridae from Heron

Island, while Sun et al. (2012) reported it from Pomacentrus amboinensis from Lizard Island. Our

discovery of this species in Chrysiptera rollandi and Pomacentrus moluccensis represent new host

records. The species has also been reported from Lizard Island, from the labrid Coris batuensis

(Bleeker, 1856) by Muñoz and Cribb (2006).

Thulinia microrchis Yamaguti, 1934

Hosts: Abudefduf septemfasciatus (Cuvier, 1830), Acanthochromis polyacanthus (Bleeker, 1855),

Dischistodus melanotus (Bleeker, 1858), Pomacentrus amboinensis Bleeker, 1868.

28



Remarks: Our specimens match the description of Bray et al. (1993a). This species has a

widespread distribution and is reported to be highly euryxenic (Bray et al. 1993a; Miller et al.

2011b). It has been reported from Heron Island in 15 species of Pomacentridae as well as in species

of Acanthuridae, Chaetodontidae, Kyphosidae, Lethrinidae, Pomacanthidae, Serranidae and

Siganidae (Yamaguti 1938; Manter 1969a; Ichihara 1986; Lester and Sewell 1990; Bray et al.

1993a; Barker et al. 1994). Despite the disparity in dietary preferences of these fishes, unpublished

molecular evidence indicates the conspecificity of this species, at least across several host families

(Miller et al. 2011b). Our finding of T. microrchis in Abudefduf septemfasciatus represents a new

host record; T. microrchis was previously reported from three other species of Abudefduf, A.

bengalensis, A. sexfasciatus and A. whitleyi (see Bray et al. 1993a).

Lepocreadiidae Odhner, 1905

Lepocreadium oyabitcha Machida, 1984

Hosts: Abudefduf bengalensis (Bloch, 1787), Abudefduf sexfasciatus (Lacépède, 1801), Abudefduf

whitleyi Allen & Robertson, 1974.



Remarks: This species was described by Machida (1984) from Abudefduf vaigiensis (Quoy &

Gaimard, 1825) from Okinawa, Japan. A single specimen was reported from A. whitleyi from

Lizard Island by Bray and Cribb (1998), who noted several differences between their specimens and

those of Machida (1984), including in body shape, having a tri-lobed (instead of ovoid) ovary, and

in infection site (the gut, as opposed to the gall-bladder). Despite this, Bray and Cribb (1998) regard

the two as being conspecific, pointing out the distinctive “ramifying moniliform-vitellarium” as an

important unifying feature. Our findings agree with those of Bray and Cribb (1998), and expand the

host range of this species by two (Abudefduf bengalensis and A. sexfasciatus). It seems likely that

this species is restricted to fishes of the genus Abudefduf.

Lepotrema adlardi (Bray, Cribb & Barker, 1993)

Host: Abudefduf bengalensis (Bloch, 1787).



Remarks: This species was described from Abudefduf bengalensis from Heron Island by Bray et al.

(1993b) (as Lepocreadium adlardi). Revision of the Lepocreadiidae by Bray and Cribb (1996) led

29

to the transfer of this species to Lepotrema. All observed features in our material are consistent with

those of L. adlardi. The species differs from other species of Lepotrema in the length of its

forebody, the extent of its vitelline follicles and having a relatively long prepharynx. Abudefduf

bengalensis remains the only known host of L. adlardi, despite the examination of several other

species of Abudefduf at both Heron and Lizard Island.

Lepotrema clavatum Ozaki, 1932

Hosts: Acanthochromis polyacanthus (Bleeker, 1855), Pomacentrus amboinensis Bleeker, 1868.



Remarks: This species was described by Ozaki (1932) from the monocanthid Stephanolepis

cirrhifer (Temminck & Schlegel, 1850) from Japan. It has since been reported in four fish families

from across the Indo-Pacific (Balistidae, Bothidae, Chaetodontidae and Pomacentridae) (see Bray

and Cribb 1996 for a full list of infected fish species). This species has been recorded in two species

of pomacentrids (Acanthochromis polyacanthus and Parma polylepis Günther, 1862) at Heron

Island on the GBR. It has also been recorded from another pomacentrid species in Hawaii,

Dascyllus albisella Gill, 1862, (Pritchard 1963). The specimens we collected are consistent with

those reported by Bray et al. (1993b) from Heron Island (as Lepocreadium clavatum), with

Pomacentrus amboinensis representing a new host record for this species. Lepotrema clavatum is

easily distinguished from the other two species of Lepotrema which infect pomacentrids. From L.

adlardi it differs in the extent of the vitelline follicles, the length of the prepharynx and length of

the forebody, and from L. monile Bray & Cribb, 1998 it differs in having a tri-lobed ovary and

lacking a sphincter around the distal metraterm. The host-specificity of L. clavatum is unusual for

this genus, being present in several fish families. Research presently underway suggests that L.

clavatum will prove to be a species complex comprising species with narrower specificity than

currently reported. This form from pomacentrids may ultimately became known under a separate

name. Records for the Great Barrier Reef show that this species is abundant in A. polyacanthus

(This study; Bray et al. 1993b) and at best uncommon in all other pomacentrids.

Lepotrema monile Bray & Cribb, 1998

Hosts: Abudefduf septemfasciatus (Cuvier, 1830), Pomacentrus amboinensis Bleeker, 1868,

Pomacentrus chrysurus Cuvier, 1830, Pomacentrus wardi Whitley, 1927.



30

Remarks: This species was first reported as Lepocreadium sp. from Pomacentrus wardi from Heron

Island (Bray et al. 1993b). Bray and Cribb (1996) revived the genus Lepotrema and included five

species within it, defined by having a combination of a dorsally-positioned excretory pore, a tri-

lobed ovary and a “folded muscular pad” on the distal surface of the metraterm. Bray and Cribb

(1998) proposed L. monile for this form, noting that while it lacks a tri-lobed ovary and the folded

muscular pad, it does possess a dorsal excretory pore. This species was subsequently reported from

Pomacentrus amboinensis from Lizard Island by Sun et al. (2012). Our specimens agree closely

with the original description of this species. Abudefduf septemfasciatus and P. chrysurus represent

new hosts for this species.

Preptetos cannoni Barker, Bray and Cribb, 1993

Hosts: Abudefduf septemfasciatus (Cuvier, 1830).


Deposited specimen: QM XXXXX

Remarks: This species was described by Barker et al. (1993) from three species of Siganidae from

Heron Island. It has also been reported from Pomacentrus bankanensis (see Bray et al. 1993b).

Species of the genus Preptetos resemble Lepocreadium Stossich, 1903, but can be distinguished by

the excretory vesicle passing between the testes. The specimen from A. septemfasciatus agrees with

the original descriptions by Barker et al. (1993) and represents a new host record for this species. In

this study, only a single individual of P. cannoni was found. The low prevalence and intensity

detected is consistent with the findings of Bray et al. (1993b) who also reported just a single

individual pomacentrid (Pomacentrus bankanensis) being infected. It seems that the species

overwhelmingly infects siganids, and occurs in pomacentrids only incidentally.

Monorchiidae Odhner, 1911

Hurleytrematoides morandi McNamara & Cribb, 2011

Host: Pomacentrus wardi Whitley, 1927.



Remarks: Species of Hurleytrematoides have been reported from several locations including Florida

(Manter 1942), China (Wang and Wang 1993), Indonesia (Machida 2005) and Australia

(McNamara et al. 2012). They overwhelmingly infect fishes from the family Chaetodontidae, with

some records from Acanthuridae, Pomacanthidae and Tetraodontidae (McNamara and Cribb 2011).

This species can be distinguished from other species of Hurleytrematoides by its distinctive body

shape and by its distinctively spined cirrus-sac (McNamara and Cribb 2011). This record, based on

31

the discovery of a single specimen in Pomacentrus wardi, is the first of this family from the

Pomacentridae. This species has previously been found only (and repeatedly) in species of

Chaetodon. This individual infection is best considered a straggler or ‘accidental’.

Transversotrematidae Witenberg, 1944

Transversotrema damsella Hunter & Cribb, 2012

Hosts: Abudefduf septemfasciatus (Cuvier, 1830), Abudefduf sexfasciatus (Lacépède, 1801).


Deposited specimens: QM XXXXX

Remarks: This species was described from Abudefduf bengalensis from Lizard Island by Hunter and

Cribb (2012) and has also been recorded from several other pomacentrid species: Abudefduf

septemfasciatus, A. sexfasciatus, Acanthochromis polyacanthus and Amblyglyphidodon curacao, all

from Lizard Island. Further examination of the host records of T. damsella suggests that this species

predominantly infects damselfishes of the genus Abudefduf (see Hunter and Cribb 2012).

Interestingly, T. damsella has not been found at Heron Island despite numerous damselfish species

having been examined there (Barker et al. 1994). Transversotrema damsella is distinguished from

other Transversotrema species by the ovary being larger than the testes and also has a relatively

distinctive overall shape.

Zoogonidae Odhner, 1902

Steganoderma gibsoni Cribb, Bray & Barker, 1992

Hosts: Abudefduf septemfasciatus (Cuvier, 1830), Abudefduf sexfasciatus (Lacépède, 1801),

Abudefduf whitleyi Allen & Robertson, 1974.



Remarks: The present specimens agree well with S. gibsoni as described originally from A. whitleyi

from the southern GBR (Cribb et al. 1992) and it was found in two further Abudefduf species. This

is the only zoogonid known from a GBR pomacentrid. The records reported here suggest that this

species is probably restricted to species of Abudefduf. There have been several other records of

zoogonids from pomacentrids; Deretrema fusillus Linton, 1910 from Abudefduf saxatilis (Linnaeus,

1758) from Florida (Linton 1910; Manter 1934,1947), Deretrema sp. from A. sordidus (Forsskål,

1775) from Saipan in Micronesia (Toman and Kamegai 1974) and Diphterostomum americanum

Manter, 1947 from Stegastes leucostictus (Müller & Troschel, 1848) from Florida (Overstreet

1969).

32

Table 1 Host-parasite associations for pomacentrid fishes and digenean trematodes from Lizard Island. Numbers indicate number of worms: numbers

in parentheses indicate number of hosts infected. Key to parasites. Derogenidae: A Derogenes pearsoni. Lecithasteridae: B, Aponurus languncula; C,

Hysterolecitha heronensis; D, Hysterolecitha nahaensis; E, Lecithaster stellatus; F, Thulinia microrchis. Lepocreadiidae: G, Lepocreadium oyabitcha;

H, Lepotrema adlardi; I, Lepotrema clavatum; J, Lepotrema monile; K, Preptetos cannoni. Bivesiculidae: L, Bivesicula unexpecta. Faustulidae: M,

Pseudobacciger cheneyae. Haplosplanchnidae: N, Schikhobalotrema sp. Zoogonidae: O, Steganoderma gibsoni. Transversotrematidae: P,

Transversotrema damsella. Monorchiidae: Q, Hurleytrematoides morandi. Fellodistomidae: R, Fellodistomidae n. sp. Gyliauchenidae: S,

Gyliauchenidae n. sp.

Parasite species

Hosts # Host A B C D E F G H I J K L M N O P Q R S

Abudefduf

bengalensis 3 4 (1) 5 (1) 2 (2)

septemfasciatus 1 3 (1) 1 (1) 1 (1) 2 (1)

sexfasciatus 15 6 (3) 10 (3) 2 (2) 3 (1)

vaigensis 4

whitleyi 8 1 (1) 5 (2) 1 (1)

Acanthochromis

polyacanthus 20 3 (3) 8 (4) 18 (10) 23 (9)

Amblyglyphidodon

curacao 20 8 (5) 1 (1)

Chromis

atripectoralis 20 2 (2)

weberi 15 96 (9)

viridis 4 1 (1)

Chrysiptera

cyanea 9 7 (2)

rex 3

rollandi 9 3 (2) 1 (1)

taupoa 1 1 (1)

Dascyllus

aruanus 20

reticulatus 22 25 (11)

trimaculatus 4

33

Dischistodus

melanotus 4 3 (1) 2 (2)

pseudochrysopecilus 4

Neoglyphidodon

melas 20 48 (6)

Neopomacentrus

azysron 21 1 (1)

Plectroglyphidodon

lacrymatus 2

Pomacentrus

adelus 17 7 (1) 2 (1) 10 (5)

amboinensis 23 1 (1) 10 (6) 3 (3) 1 (1)

chrysurus 22 7 (4) 1 (1) 1 (1)

coelestis 12 1 (1)

moluccensis 27 18 (6) 2 (2) 1 (1) 2 (1)

nagasakiensis 7 3 (2)

pavo 5 1 (1)

wardi 8 4 (3) 1 (1)

Stegastes

apicalis 7 8 (2) 1 (1) 7 (3)

fasciolatus 1

34

Discussion

Identification of the fauna

From a total of 358 individuals of 32 species of pomacentrids examined we found 19 species of

digenean trematodes and 54 host/parasite combinations. A total of 19 host distributions were new,

18 were new host records, and three were new species. The characterisation of this fauna proved in

part straightforward and in part complex. It was straightforward in that the great majority of the

species were easily identified as forms previously reported from pomacentrids or, in a few cases,

other fishes. The three new species, a faustulid, a fellodistomid and a gyliauchenid, were all readily

recognisable as distinct and were the sole representatives of their respective families encountered.

The complexities of identification relate almost entirely to the delineation of the Hysterolecitha

nahaensis complex. As outlined above, this species was by far the most common infection

encountered but combined molecular and morphological evidence suggests that it comprises at least

four species, there is limited evidence for as many as seven species, and it seems entirely possible

that there could be more than that. On the basis of the present evidence it is possible to distinguish

only three taxa by morphology of which the most common, Morphotype 3, almost certainly is a

complex of multiple species. In addition to the problem relative to H. nahaensis, it seems likely that

some of the other species identified (especially Aponurus languncula, Lecithaster stellatus and

Lepotrema clavatum) will ultimately become recognised under different names, but their distinction

as single taxa in pomacentrids does not appear to be in doubt.

The second complexity in the characterisation of this fauna is in the size of the task. The

study incorporated the examination of 372 individuals of 32 pomacentrid species. Despite the fact

that these numbers represent a substantial number of individual fishes for field and processing

effort, the incompleteness of the survey of the fauna is demonstrated by two features of the data.

First, there remain a further 58 pomacentrid species known from Lizard Island that have never been

examined for parasites; undoubtedly these will harbour many new host-parasite combinations. The

second feature of the fauna is the rarity of many infections. Of 54 host/parasite combinations, 27

(exactly half) were detected in only a single individual fish. Such a level of rarity implies that a

great many more host/parasite combinations remain undetected for the hosts that have been

examined. Some of the infections found only once are evidently stragglers (e.g. Preptetos cannoni

and Hurleytrematoides morandi) and are thus insignificant in the understanding of the fauna. In

contrast, the distribution of oioxenic (single host) species is difficult to predict and there are likely

to be a number of those that remain completely undetected so far.

Below we consider the nature of host-specificity, overall richness patterns and beta diversity

for this fauna with the understanding that the data set remains incomplete.

35

Richness

Despite the Pomacentridae being a large and diverse group of coral reef fishes, the mean

species richness of digeneans in species within this family was low at 1.69. Results show that 20 of

32 pomacentrid species examined from Lizard Island were infected with only 1 ̶ 3 species of

digeneans; just five species harboured four species of digenean; seven of the 32 species of

pomacentrids had no digeneans at all. For many pomacentrid species the samples were low, but it is

striking that the 20 Dascyllus aruanus produced no infections at all.

It is well-established that parasite richness within host species is not random, but is broadly

related to host species traits, especially size, density and distribution (Guégan et al. 1992; Kamiya et

al. 2014). Host size probably explains the relatively low richness of adult digeneans among

pomacentrid species relative to larger coral reef fishes such as chaetodontids, lutjanids and serranids

as well as among the pomacentrid species. Notably the larger species in the study, species of

Abudefduf, were infected with most digenean species. Generally, host size is a strong predictor for

parasite richness because larger-bodied individuals provide greater space and resources for parasites

than smaller-bodied individuals (Sasal et al. 1997; Kamiya et al. 2014). However, at the

infracommunity level, parasite richness is more variable when examining the effects of increasing

host size. Two medium-sized pomacentrid species, Pomacentrus amboinensis and P. moluccensis,

were also infected with 4 species of digeneans; notably, however these two species had the two

largest samples examined in this study indicating the importance of sample size. Overall, although

host size is a good predictor of parasite richness it will not necessarily over-rule the effects of

different habitats, densities and diet (Lo et al. 1998; Torres et al. 2006; Morand 2015).

The reported digenean fauna of pomacentrids from both the northern and southern GBR

suggests that there has been little radiation of families of trematodes species within this group; the

total of 23 species are drawn from 11 trematode families. Apart from the presence of three species

of Lepotrema the clear exception is the H. nahaensis complex; however, interpretation of the nature

of the H. nahaensis complex must await its better characterisation. The limited radiation of

individual genera is in contrast to many other GBR fish family/trematode family complexes

Bucephalidae Poche, 1907 in serranids (Bott and Cribb 2009), Paradiscogaster Yamaguti, 1934

(Diaz et al. 2013) and Hurleytrematoides Yamaguti,1953 (McNamara and Cribb 2011) in

chaetodontids, Opecoelidae Ozaki, 1925 in mullids (Rohner and Cribb 2013) and Cryptogonimidae

Ward, 1917 in lutjanids (Miller and Cribb 2007). In all these combinations there is much greater

parasite richness than seen for any of the families represented in pomacentrids.

As found by Barker et al. (1994) at Heron Island, the diversity of pomacentrid species is far

greater than the diversity of digenean trematodes at Lizard Island. Other studies from other

locations have also shown that the diversity of digeneans in pomacentrids is relatively low (Yeo and

36

Spieler 1980; Dyer et al. 1985). However, it should be noted that this study only examined 32

species of pomacentrids from a possible 89 species that have been recorded at Lizard Island

(Grutter et al. in review). Thus, there is the likelihood that other species of digenean worms will be

found, as well as extending the host records for already known species.

Host-specificity

The host specificity of the digeneans of pomacentrids varies, with a tendency towards being

relatively low. We found just three1 species to be convincingly oioxenic (i.e found repeatedly in

only one host species) at Lizard Island: Bivesicula unexpecta was only found in Acanthochromis

polyacanthus, Lepotrema adlardi only in Abudefduf bengalensis and Pseudobacciger cheneyae only

in Chromis weberi. Due to only a single specimen being recorded, further examination is required

to determine whether Fellodistomidae sp. from Chrysiptera rollandi is indeed oioxenic or in fact

occurs in other pomacentrids. Our findings are in support of earlier studies regarding Bivesicula

unexpecta and Lepotrema adlardi (Bray et al. 1993b; Cribb et al. 1994a). The majority of the

digenean species reported in this study are apparently stenoxenic, i.e. infecting multiple species

within the same family; Derogenes pearsoni, Hysterolecitha heronensis, H. nahaensis complex,

Lepocreadium oyabitcha, Lepotrema monile, Steganoderma gibsoni (see Cribb et al. 1992; Bray et

al. 1993a; Bray and Cribb 1998) and Transversotrema damsella (Hunter and Cribb 2012) are all in

this category. Two of these stenoxenic species, H. heronensis and L. oyabitcha, can be further

categorised as being restricted to multiple congeners of Pomacentrus and Abudefduf, respectively.

The remaining species are classified as euryxenic, infecting a wide range of hosts from multiple

families; Aponurus laguncula, Lecithaster stellatus, Lepotrema clavatum, Hurleytrematoides

morandi, Preptetos cannoni and Thulinia microrchis. However, it should be noted that H. morandi

and P. cannoni, which are common in chaetodontids and siganids respectively are technically

euryxenic, but are effectively stenoxenic to the Chaetodontidae and Siganidae, infecting

pomacentrids only incidentally. This could also be the case for A. languncula which has not been

reported other than from three species of pomacentrids. Infections of these pomacentrid species are

rare, suggesting that the infections may also be incidental.

A likely explanation for the generally broad host range of many species are the eco-

physiological similarities between pomacentrid species, especially their broad and overlapping

diets, which may facilitate host-sharing in digeneans (Marcogliese 2002). Use of intermediate hosts

by metacercariae facilitates transmission into the definitive host (Poulin and Cribb 2002). Broadly

overlapping diets increase the possibility that multiple species of fish will encounter the same suite

1 In this study the gyliauchenid was oioxenic, but subsequent collections by colleagues (not reported

here) has shown that it infects at least two species in addition to Stegastes apicalis.

37

of metacercariae. The fact that hemiuroid metacercariae are known to infect a range of zooplankton

such as calanoid copepods and polychaetes (Marcogliese 1995) which in turn are consumed by

many pomacentrids, may explain the high prevalence of digeneans from this superfamily infecting

multiple pomacentrid species. Although diet may explain some of the variation in the infection of

pomacentrids by certain digeneans species (e.g. Hysterolecitha nahaensis complex), it does not

explain the high specificity observed in other species (e.g. Pseudobacciger cheneyae to just one

species of Chromis, Lepotrema adlardi to just one species of Abudefduf). In such cases the

explanation may be physiological compatibility (Adamson and Caira 1994). The physiological

compatibility between a parasite and its potential hosts may play an important role in determining

the suitability of those hosts, thus driving host specificity. However, the extent to which physiology

influences host specificity has not been explored for most fish-infecting digeneans and is poorly

understood (Miller et al. 2011b).

It is notable that some of the digeneans we found were species which are interpreted as

having cosmopolitan distributions and exhibiting euryxenicity, i.e. extremely wide host ranges. We

encountered five species which could be classed as being euryxenic, including two (Aponurus

languncula and Lecithaster stellatus) that are reported to be distributed circumglobally.

Euryxenicity is observed in a small number of marine digenean species worldwide (Bartoli et al.

2005) and has been reported for only 23 of the 290 species of digeneans on the GBR (Miller et al.

2011b). It is possible, perhaps probable, that some of these euryxenic species may comprise

complexes of cryptic species that have distinct geographical distributions. The two cosmopolitan

species reported in this study are from the superfamily Hemiuroidea, which are characterised by

many of its member taxa exhibiting low host specificity (Gibson and Bray 1979).

Although A. laguncula has not been reported from any fish family on the GBR other than

pomacentrids (reported from 3 species only), it has been recorded in 67 teleost species from 29

families, making it putatively one of the most widespread digeneans with respect to both

geographical distribution and host range (Bray and MacKenzie 1990; Bray et al. 1993a). It is

currently unclear whether all records of A. laguncula comprise one species or form a species

complex. Recent molecular data have indicated the presence of a cryptic species of Aponurus within

the A. laguncula complex, A. mulli Carreras-Aubets, Repulles-Albelda, Kostadinova & Carrasson,

2011, which infects red mullet, Mullus barbatus Linnaeus, 1758, from the Mediterranean (Carreras-

Aubets et al. 2011). It is possible that further exploration will reveal that Aponurus laguncula is

actually comprised of multiple cryptic species, and that A. laguncula sensu stricto may actually not

be found in Australian waters. Conversely, it is also possible that molecular work may reveal that A.

laguncula is indeed one species, despite its apparent cosmopolitan distribution, as has been

previously demonstrated for the tuna-infecting aporocotylid fluke Cardicola forsteri Cribb, Daintith

38

& Munday, 2000 (Aiken et al. 2007). The same principles may also apply to the other reportedly

cosmopolitan species, Lecithaster stellatus.

Cryptic species may also occur within some of the stenoxenic species found on the GBR, in

particular Hysterolecitha nahaensis. Currently, H. nahaensis has been recorded in a total of 26

pomacentrid species from Lizard and Heron Islands, making it the digenean with the widest host

range. This species closely resembles H. heronensis; the two species may have been historically

confused with each other due to their morphological similarities (Bray et al. 1993a). These species

require comprehensive reconsideration with combined morphological or molecular analysis. It is

apparent that a number of cryptic species are present within these species, as has been demonstrated

for other digeneans in other hosts (Jousson et al. 2000; Nolan and Cribb 2005; Diaz et al. 2015).

Beta diversity

These data create a rare opportunity to consider beta diversity in parasite faunal composition on the

GBR. Both the depth of study and richness of digenean trematodes encountered in this study is

comparable to that of Barker et al. (1994), who examined 312 individuals from 39 species of

pomacentrids from Heron Island, on the southern GBR (see Table 2). Twenty-one pomacentrid

species were examined at both sites. Of these, six had robust samples of at least 10 individuals at

both sites. Although the overall trematode richness is almost identical for the two sites, it is striking

that mean richness and prevalence were reported as much higher at Heron than at Lizard Island.

Table 2 Comparison between data collected from Heron Island (Barker et al. 1994) and Lizard

Island (present study) for digeneans from pomacentrid fishes.

HI LI

Fish species 39 32

Fish individuals 312 358

Parasite species 18 19

Host/parasite combinations 86 54

Mean richness 2.205 1.688

Total infections 209 140

Mean overall prevalence 0.67 0.39

39

The total digenean fauna for GBR pomacentrids now stands at 23 species in 50 pomacentrid

species of which 14 are now known from both locations, five have been only recorded from Lizard

Island (Lepocreadium oyabitcha, Pseudobacciger cheneyae, Transversotrema damsella,

Fellodistomid sp and Gyliauchenid sp.) and four only from Heron Island (Derogenes pharyngicola,

Lepocreadium sp., Preptetos xesuri and Macvicaria heronensis). Of the five species known only

from Lizard Island, only three can be inferred to be absent from Heron Island. As only a single

specimen of the fellodistomid digenean was recorded and given that no Chromis weberi have been

examined at Heron Island for P. cheneyae, despite records indicating that C. weberi does occur

there (Russell 1983), we cannot conclude that these species do not occur at Lizard Island; it is

possible that with further examination, these species may prove to be present at Heron Island. In

contrast, the evidence is good that three species (L. oyabitcha, T. damsella and Gyliauchenid n. sp.)

do not occur at Heron Island. Of the four species known only from Heron Island, only D.

pharyngicola is a regularly occurring parasite restricted to pomacentrids; Preptetos xesuri and

Macvicaria heronensis are mainly parasites of other fishes and the Lepocreadium sp. from

Amblyglyphidodon curacao has been collected only twice. We conclude that it is possible to

identify a “core pomacentrid fauna” of species that are regular in and dependent on pomacentrids

(Table 3). From this core fauna are excluded species that are rare to the point that their status is

uncertain (e.g. Aponurus laguncula, Fellodistomid sp.) and those that are clearly principally

parasites of other fishes (e.g. Preptetos cannoni and Hurleytrematoides morandi). With such

exclusions we recognise 14 morphospecies in this category with the understanding that “H.

nahaensis complex” certainly represents multiple species.

40

Table 3 Core pomacentrid specific digeneans from Heron and Lizard Island.

The only previous detailed comparison of the parasite fauna of comparable fish from the

southern and northern GBR was that of monorchiid trematodes of chaetodontid fishes (McNamara

and Cribb 2011; McNamara et al. 2012). A total of 10 species in total was reported, of which all

occurred at Heron Island but only seven were found at Lizard Island. This finding ran counter to the

general pattern of greater biological richness on the northern than on the southern GBR (Russell

1983). Here the comparison is less straightforward because of the complex (and partly uncertain)

host-specificity of the pomacentrid parasites (whereas the monorchiids of chaetodontids show

reliably high specificity) and because a smaller number of the host species sampled at the two sites

were in common. Of the 14 core pomacentrid species, 10 (71 %) have been found at Heron Island

and 13 (93 %) at Lizard Island (Table 3). These numbers are consistent with the interpretation that

there are significant differences in the parasite fauna at the two sites and that overall richness is

slightly higher in the north. That both mean species richness and mean overall prevalence were

reported as much higher at Heron Island by Barker et al. (1994) is puzzling. There are so many

areas of uncertainty in these comparisons (from the true identity of several of the species to the

robustness of the sample sizes) that these findings can only be a basis for future analyses.

Family Genus Species HI LI

Bivesiculidae Bivesicula unexpecta ● ●

Derogenidae Derogenes pearsoni ● ●

Derogenes pharyngicola ●

Faustulidae Pseudobacciger cheneyae ●

Gyliauchenidae Gyliauchen sp. ●

Haplosplanchnidae Schickhobalotrema pomacentri ● ●

Lecithasteridae Hysterolecitha heronensis ● ●

Hysterolecitha nahaensis

(complex) ● ●

Lepocreadiidae Lepocreadium oyabitcha ●

Lepotrema adlardi ● ●

Lepotrema clavatum ● ●

Lepotrema monile ● ●

Transversotrematidae Transversotrema damsella ●

Zoogonidae Steganoderma gibsoni ● ●

Count 10 13

41

Acknowledgments

We thank E. C. McClure, F. Cortesi, K. L. Cheney, S. Martin for their help collecting specimens

and the staff of the Lizard Island Research Station for their support. This work was funded by a Sea

World Research and Rescue Foundation, Australia (SWRRFI) awarded to D. Sun and T. H. Cribb

and an Australian Research Council Discovery Grant awarded to T. H. Cribb.

42

Pseudobacciger cheneyae n. sp. (Digenea:

Gymnophalloidea) from Weber’s chromis (Chromis

weberi Fowler & Bean) (Perciformes: Pomacentridae) at

Lizard Island, Great Barrier Reef, Australia

43

Pseudobacciger cheneyae n. sp. (Digenea: Gymnophalloidea) from Weber’s

chromis (Chromis weberi Fowler & Bean) (Perciformes: Pomacentridae) at

Lizard Island, Great Barrier Reef, Australia

Derek Sun, Rodney A. Bray, Russell Q-Y. Yong, Scott C. Cutmore & Thomas H. Cribb

This section is published as an original article in Systematic Parasitology (2014) 8:141-152

Abstract

A new species of digenean, Pseudobacciger cheneyae n. sp., is described from the intestines of

Weber’s chromis (Chromis weberi Fowler & Bean) from off Lizard Island, Great Barrier Reef,

Australia. This species differs from the three described species of Pseudobacciger Nahhas & Cable,

1964 [P. cablei Madhavi, 1975, P. harengulae (Yamaguti, 1938) and P. manteri Nahhas & Cable,

1964], in combinations of the size of the suckers and the length of the caeca. The host of the present

species is a perciform (Family Pomacentridae) which contrasts with previous records of the genus

which are almost exclusively from clupeiform fishes. The genus Pseudobacciger is presently

recognised within the family Faustulidae (Poche, 1926) but phylogenetic analyses of 28S and ITS2

rDNA show that the new species bears no relationship to species of four other faustulid genera

(Antorchis Linton, 1911, Bacciger Nicoll, 1924, Paradiscogaster Yamaguti, 1934 and

Trigonocryptus Martin, 1958) but that instead it is nested within the Gymnophalloidea (Odhner,

1905) as sister to the Tandanicolidae (Johnston, 1927). This result suggests that the Faustulidae is

polyphyletic.

Introduction

The Faustulidae Poche, 1926 is a small family of digenean trematodes found in a wide range of

mainly marine teleost families (Bray 2008). There are 11 genera, most of which have few species.

Only four genera, Antorchis Linton, 1911, Bacciger Nicoll, 1924, Faustula Poche, 1926 and

Paradiscogaster Yamaguti, 1934, have five or more described species. In Australia, faustulids have

been reported from several fish families in tropical, sub-tropical and temperate waters (Korotaeva

1969; Kurochkin 1970; Bray 1982; Bray et al. 1994; Cribb et al. 1999; Diaz and Cribb 2013; Diaz

et al. 2013).

44

Pseudobacciger Nahhas & Cable, 1964, as reviewed by Dimitrov et al. (1999), is

represented by three species, all infecting marine teleosts. The hosts are overwhelmingly

clupeiforms (Clupeidae and Engraulidae), with single records from a sparid and a pomacanthid

(Madhavi 1975; Parukhin 1976a; Gaevskaja and Naidenova 1996; Dimitrov et al. 1999; Machida

and Uchida 2001). Species of Pseudobacciger primarily infect the intestine, but also occur

occasionally in the pyloric caeca (Dimitrov et al. 1999). There are reports of species of

Pseudobacciger from the temperate to subtropical Atlantic, Pacific and Indian Oceans (Bray 2008)

but to date there are no records from Australian waters. Sporocysts have been reported in bivalves

and metacercariae in shrimps from Japan and Korea (Chun et al. 1981; Kim and Chun 1984).

Pseudobacciger was previously included in the subfamily Baccigerinae Yamaguti, 1958 (family

Fellodistomidae Nicoll, 1909) but molecular analysis by Hall et al. (1999) revealed the

Fellodistomidae to be polyphyletic, and the Baccigerinae was raised to family level as its senior

synonym, Faustulidae.

Pomacentrids have been extensively examined for trematodes in Australia (Manter 1969b;

Cribb et al. 1992; Bray et al. 1993a; Bray et al. 1993b; Bray and Cribb 1998; Bray and Cribb 2004;

Hunter and Cribb 2012). To date 20 fully identified species from eight families of trematodes have

been reported, but no faustulids have been recorded from pomacentrids from Australia or

elsewhere. Between July 2011 and August 2012, several specimens of a planktivorous damselfish,

Weber’s chromis (Chromis weberi Fowler & Bean), were caught from off Lizard Island, on the

northern GBR and examined for trematodes. In these we discovered a previously unreported

trematode, of which morphological analysis suggests is a new species of Pseudobacciger.

Molecular phylogenetic analysis suggests that the family-level affiliation of this new species is not

clear.


Specimen collection

Fifteen Chromis weberi were collected from off Lizard Island, Great Barrier Reef, Australia in 2011

and 2012. Fish were collected using a barrier net and were killed prior to dissection by cranial

pithing. The contents of the body cavity were removed and examined under a dissecting microscope

using fine scissors and forceps. The gastrointestinal tract was examined for parasites using the gut-

wash approach as described by Cribb and Bray (2010). Specimens were fixed by pipetting them into

near boiling saline followed by immediate preservation in 70 % ethanol for both morphological and

molecular analysis. Worms for morphological analysis were stained using Mayer’s haematoxylin,

destained in 1 % hydrochloric acid, neutralized with 0.5 % ammonia solution, dehydrated through a

45

graded series of ethanol, cleared in methyl salicylate, and mounted on slides in Canada balsam.

Drawings were made with the aid of a camera lucida. Measurements were made using a SPOT

Insight™ digital camera (Diagnostic Instruments, Inc.) mounted on an Olympus BH-2 compound

microscope using SPOT™ imaging software. Measurements are in micrometres (µm) and are

presented as the range, followed by the mean in parentheses. Type-specimens are lodged in the

Queensland Museum (QM), Brisbane, Australia.

Molecular analysis

All voucher specimens are lodged in the Queensland Museum, Brisbane, Australia and are

paragenophores (sensu Pleijel et al. 2008). Total genomic DNA was extracted using standard

phenol/chloroform extraction techniques (Sambrook and Russell 2001). Partial 28S nuclear

ribosomal DNA was amplified using the primers LSU5 (5'-TAG GTC GAC CCG CTG AAY TTA

AGC-3') (Littlewood et al. 2000) and ECD2 (5'-CTT GGT CCG TGT TTC AAG ACG GG-3')

(Littlewood et al. 2000) and the ITS2 region using the primers 3S (5'-GGT ACC GGT GGA TCA

CGT GGC TAG TG-3') (Bowles et al. 1993) and ITS2.2 (5'-CCT GGT TAG TTT CTT TTC CTC

CGC-3' ) (Cribb et al. 1998). PCR for both the 28S and ITS2 regions was performed with a total

volume of 20 μl consisting of approximately 10 ng of DNA, 5 μl of 5 MyTaq Reaction Buffer

(Bioline), 0.75 μl of each primer (10 pmols) and 0.25 μl of Taq DNA polymerase (Bioline

MyTaq™ DNA Polymerase), made up to 20 μl with Invitrogen™ ultraPURE™ distilled water.

Amplification was carried out on a MJ Research PTC-150 thermocycler. The following profile was

used to amplify the 28S region: an initial 95°C denaturation for 4 min, followed by 30 cycles of

95°C denaturation for 1 min, 56°C annealing for 1 min, 72°C extension for 2 min, followed by a

single cycle of 95°C denaturation for 1 min, 55°C annealing for 45 sec and a final 72°C extension

for 4 min. The following profile was used to amplify the ITS2 region: an initial single cycle of 95°C

denaturation for 3 min, 45°C annealing for 2 min, 72°C extension for 90 sec, followed by 4 cycles

of 95°C denaturation for 45 sec, 50°C annealing for 45 sec, 72°C extension for 90 sec, followed by

30 cycles of 95°C denaturation for 20 sec, 52°C annealing for 20 sec, 72°C extension for 90 sec,

followed by a final 72°C extension for 5 min. Amplified DNA was purified using a Bioline

ISOLATE II PCR and Gel Kit, according to the manufacturer’s protocol. Cycle sequencing of

purified DNA was carried out for the 28S rDNA region using the same primers used for PCR

amplification as well as the additional primer 300F (5'-CAA GTA CCG TGA GGG AAA GTT-3')

(Littlewood et al. 2000) and for the ITS2 rDNA region using the forward primer GA1 (5'-AGA

ACA TCG ACA TCT TGA AC-3') (Anderson and Barker 1998) and the reverse primer ITS2.2.

Cycle sequencing was carried out at the Australian Genome Research Facility using an AB3730xl

capillary sequencer. Sequencher™ version 4.5 (GeneCodes Corp.) was used to assemble and edit

46

contiguous sequences. GenBank accession numbers for the taxa sequenced in this study are shown

in Table 1.

In addition to samples of the species under direct consideration, we generated 28S and ITS2

rDNA sequences for an unknown gymnophallid metacercaria collected from the bivalve Paphies

elongata (Reeve) (Mesodesmatidae) collected from Woorim Beach, Bribie Island, southeast

Queensland (27°04'S, 153°12'E). Collection of this specimen was required as there are currently no

28S sequence available on GenBank for a gymnophallid. Voucher specimens of this species are

lodged within the Queensland Museum under registration numbers QM G 234368 – G 234387. We

also generated an ITS2 rDNA sequence from a specimen of Tandanicola bancrofti Johnston, 1927

consistent with the original description of this species (Johnston 1927).

Table 1 Collection data and GenBank accession numbers for taxa sequenced during this study

Taxon Host Locality ITS2 rDNA

(no. of

replicates)

28S rDNA

(no. of

replicates)

Pseudobacciger

cheneyae n. sp.

Chromis weberi

Fowler & Bean

Off Lizard Island, Great

Barrier Reef,

Queensland (14°40′S,

145°27′E)

KJ648920 (2) KJ648919 (2)

Gymnophallidae

sp.

Paphies elongata

(Reeve)

Woorim Beach, Bribie

Island, Queensland

(27°04'S, 153°12'E)

KJ648918 (2) KJ648917 (1)

Tandanicola

bancrofti

(Johnston,

1927)

Tandanus tandanus

(Mitchell)

Kholo Crossing,

Brisbane River,

Queensland (27°33'S,

152°44'E)

KJ648921 (2)

Phylogenetic analysis

The ITS2 and partial 28S rDNA sequences generated during this study were aligned with species of

Bucephalidae Poche, 1907, Faustulidae, Fellodistomidae and Tandanicolidae Johnston, 1927

available on GenBank using MUSCLE version 3.7 (Edgar 2004) with ClustalW sequence weighting

and UPGMA clustering for iterations 1 and 2. The resultant alignments were refined by eye using

MESQUITE (Maddison and Maddison 2009) and the ends of each fragment were trimmed to match

the shortest sequence in the alignments.

Bayesian inference and Maximum Likelihood analyses of both the 28S and ITS2 rDNA

datasets were conducted to explore relationships among these taxa. Bayesian inference analyses

were performed using MrBayes version 3.2.2 (Ronquist et al. 2012), run on the CIPRES portal

47

(Miller et al. 2010). Maximum Likelihood analyses were performed using RAxML version 7.2.8

(Stamatakis et al. 2008). The software jModelTest version 0.1.1 (Posada 2008) was used to estimate

the best nucleotide substitution model for the datasets. Bayesian inference analyses were conducted

on the 28S dataset using the GTR+G model and on the ITS2 dataset using the TVM+G model

predicted as the best estimators by the Akaike Information Criterion (AIC) in jModelTest. The

closest approximation of these models was implemented in the subsequent Maximum Likelihood

analyses. Bayesian inference analyses were run over 10,000,000 generations (ngen = 10,000,000)

with two runs each containing four simultaneous Markov Chain Monte Carlo (MCMC) chains

(nchains = 4) and every 1,000th tree saved (samplefreq = 1,000). Bayesian analyses used the

following parameters: nst = 6, rates = gamma, ngammacat = 4, and the priors parameters of the

combined dataset were set to ratepr = variable. Samples of substitution model parameters, and tree

and branch lengths were summarised using the parameters ‘sump burnin = 3000’ and ‘sumt burnin

= 3,000’. These ‘burnin’ parameters were chosen because the log likelihood scores ‘stabilised’ well

before 300,000 replicates in the Bayesian inference analyses. Nodal support in the Maximum

Likelihood analyses was estimated by performing 100 bootstrap pseudoreplicates. Species of the

Hemiuridae (Looss, 1899) were designated as functional outgroups in all analyses.

Gymnophalloidea Odhner, 1905 incertae familiae

Pseudobacciger Nahhas & Cable, 1964

Pseudobacciger cheneyae n. sp.

Type-host: Chromis weberi Fowler & Bean (Perciformes: Pomacentridae), Weber’s chromis.

Type-locality: Off Lizard Island, northern Great Barrier Reef, Queensland, Australia (14°40'S,

145°27'E).

Site in host: Intestine.

Material examined: A total of 100 specimens collected from C. weberi; 2 for molecular analysis,

the remainder for morphological analysis.

Prevalence: 73% (11 of 15 fish infected).

Intensity: 1−44 individuals per fish with a mean ( SE) intensity (10 4.128)

Type-material: Holotype QM G 234368 and 19 paratypes QM G 234369–G 234387 deposited in

the Queensland Museum, Brisbane, Australia.

urn:lsid:zoobank.org:act: 369B1E27-B899-48CD-ACD6-0893B314C1C4 urn:lsid:zoobank.org:act: AB0A60B3-E5AC-4EE4-8990-8664A738A1E9

48

Molecular sequence data: 28S rDNA, 2 replicates (1 submitted; GenBank KJ648919); ITS2 rDNA,

2 replicates (1 submitted; GenBank KJ648920).

Etymology: This species is named in honour of Dr Karen Cheney, The University of Queensland, in

recognition of her contribution to marine parasitology.

Description (Figure 1, 2)

[Measurements taken from 20 specimens.] Body pyriform, posterior half somewhat broader than

anterior, 371453 (407) 230327 (281). Forebody, 98138 (119) long, occupies 2334 (28) %

body length. Tegumental spines minute, <1 long, cover entire body evenly, including surfaces of

oral and ventral suckers. Oral sucker well delineated, opening ventro-subterminally, muscular,

6891 4663 (79 57). Ventral sucker well developed, in anterior half of body, 80109

69101 (94 87). Oral sucker length to ventral sucker length ratio 1:1.211.67 (1.47). Oral sucker

width to ventral sucker width ratio 1: 1.031.24 (1.13). Prepharynx absent. Pharynx muscular,

immediately posterior to oral sucker, 3647 2534 (40 29). Oesophagus distinct, often sinuous

or coiling once before intestinal bifurcation, 3355 (43) long. Intestine bifurcates dorsal to ventral

sucker; caeca usually hidden by uterine coils, extend to anterior hindbody, close to posterior

margins of testes, 115166 (133) long (Figure 2). Testes 2, entire, nearly round or somewhat

irregular in outline, opposite in anterior hindbody, intercaecal, well-separated from lateral margins

of body, 4781 5181 (65 70). Sperm ducts inconspicuous. Cirrus-sac absent. Seminal vesicle

bipartite, median, posterior to caecal bifurcation, reaching anteriorly from level of posterior margin

of ventral sucker, 2671 3270 (43 47). Pars prostatica and ejaculatory ducts not detected.

Genital pore median, immediately anterior to margin of ventral sucker. Ovary trilobed, median,

between testes, posterior margin usually hidden by uterine coils, extends to anterior hindbody,

4473 5596 (54 75). Oviduct and oötype inconspicuous. Laurer’s canal opening dorsally

towards posterior end of body (usually obscured by eggs, seen clearly in only a single laterally

mounted specimen). Seminal receptacle not detected. Uterine coils extensive in posterior half of

body from posterior margin of vitelline follicles and vitelline duct, and overlapping testes, ovary

and seminal vesicle. Vitelline follicles in 2 small lateral fields at level of ventral sucker, 64100

3373 (84 47). Vitelline ducts undulate, unite medially anterior to ovary and posterior to seminal

vesicle, straight length 82115 (96), total length 89143 (115). Eggs elliptical, 1821 1215 (20

13.5). Excretory vesicle Y-shaped; arms extend to level of anterior margin of ventral sucker,

243326 (286) long. Excretory pore terminal.

49

2

Figure 1-2 Pseudobacciger cheneyae n. sp. ex Chromis weberi from off Lizard Island. 1, Holotype,

whole mount, ventral view; 2, Holotype, whole mount, ventral view, showing excretory vesicle,

caeca and seminal vesicle. Scale-bars: 100 μm

Molecular and phylogenetic analysis

BLAST analysis of the 28S sequence identified the greatest similarity of Pseudobacciger cheneyae

n. sp. as being to two tandanicolids, Tandanicola bancrofti Johnston, 1927 and Prosogonarium

angelae Cribb & Bray, 1994, rather than to faustulids as had been expected given the generic level

identification of the species. The sequence for Pseudobacciger cheneyae n. sp. was thus aligned

with and analysed relative to a wide range of trematodes of the suborder Bucephalata La Rue, 1926

(Bucephaloidea Poche, 1907: Bucephalidae; Gymnophalloidea Odhner, 1905: Fellodistomidae,

Gymnophallidae Odhner, 1905 and Tandanicolidae) and, from the suborder Xiphidiata Olson,

Cribb, Tkach, Bray & Littlewood, 2003: Faustulidae (Table 2). Figure 3 shows relationships

derived from Bayesian inference and Maximum Likelihood analyses of P. cheneyae n. sp. relative

to these taxa. In this analysis P. cheneyae n. sp. forms a strongly supported clade with the two

tandanicolids within the well-supported Gymnophalloidea and is distant from the faustulids.

1

50

BLAST analysis of the ITS2 sequence identified the greatest similarity of P. cheneyae n. sp.

as being to a range of bucephalids. Given this, and the results of the 28S rDNA analyses, the

sequence was aligned with and analysed relative to a wide range of trematodes of the suborder

Bucephalata (Bucephaloidea: Bucephalidae; Gymnophalloidea: Fellodistomidae, Gymnophallidae

and Tandanicolidae) and, as above, from the suborder Xiphidiata, representatives of Faustulidae. In

all analyses P. cheneyae n. sp. formed clades with a range of taxa from the Bucephalata; it was

never associated with the clade comprising faustulids. However, posterior probability and bootstrap

support were typically low and these analyses are not therefore shown.

Table 2 28S rDNA sequence data from GenBank included in phylogenetic analyses

Species GenBank

accession

number

Reference

Family Fellodistomidae

Coomera brayi Dove & Cribb, 1995 KJ425462 Cribb et al. (2014a)

Fellodistomum agnotum Nicoll, 1909 AJ405289 Bray et al. (1999)

Fellodistomum fellis (Olsson, 1868) AJ405290 Bray et al. (1999)

Oceroma praecox Cribb, Miller, Bray & Cutmore, 2014 KJ425464 Cribb et al. (2014a)

Olssonium turneri Bray & Gibson, 1980 AY222283 Olson et al. (2003)

Steringophorus blackeri Bray, 1973 AJ405296 Bray et al. (1999)

Steringophorus dorsolineatus (Reimer, 1985) AJ405291 Bray et al. (1999)

Steringophorus furciger (Olsson, 1867) AJ405292 Bray et al. (1999)

Steringophorus margolisi Bray, 1995 AY222281 Olson et al. (2003)

Steringophorus pritchardae (Campbell, 1975) AJ405295 Bray et al. (1999)

Steringophorus thulini Bray & Gibson, 1980 AJ405298 Bray et al. (1999)

Tergestia sp. KJ425467 Cribb et al. (2014a)

Family Tandanicolidae

Prosogonarium angelae Cribb & Bray, 1994 AY222285 Olson et al. (2003)

Tandanicola bancrofti Johnston, 1927 KJ425466 Cribb et al. (2014a)

Family Bucephalidae

Rhipidocotyle minima (Wagener, 1852)* AY222225 Olson et al. (2003)

Family Faustulidae

Antorchis pomacanthi (Hafeezullah & Siddiqi, 1970) AY222268 Olson et al. (2003)

Bacciger lesteri Bray, 1982 AY222269 Olson et al. (2003)

Trigonocryptus conus Martin, 1958 AY222270 Olson et al. (2003)

Outgroup Taxa

Superfamily Hemiuroidea

Aponurus sp. DQ354368 Pankov et al. (2006)

Bunocotyle progenetica Chabaud & Buttner, 1959 DQ354365 Pankov et al. (2006)

Robinia aurata Pankov, Webster, Blasco-Costa, Gibson,

Littlewood & Kostadinova, 2006

DQ354367 Pankov et al. (2006)

*Bartoli et al. (2006) re-identified this material as R. minima (Wagener, 1852) not R. galeata (Rudolphi,

1819) as originally identified. R. galeata, the type-species of the genus, was poorly known until redescribed

by Derbel et al. (2011), who showed that it is clearly distinct from R. minima.

51

Figure 3 Relationships between Pseudobacciger cheneyae n. sp. and other Gymnophalloidea,

Bucephalidae and Faustulidae taxa based on Bayesian inference and Maximum Likelihood analyses

of the 28S rDNA dataset. Bayesian inference posterior probabilities are shown above the nodes and

Maximum Likelihood bootstrap support values below the nodes; values < 50 not shown.

Abbreviations: Bu, Bucephalidae; Faustul, Faustulidae; Gy, Gymnophallidae; Tand,

Tandanicolidae.

Discussion

Taxonomy

The present species possesses features that are fully consistent with species of the faustulid genus

Pseudobacciger as recognised by Bray (2008). The present species has the same general

organization and, in particular, possesses the naked two-chambered seminal vesicle reported for that

genus. As reviewed by Dimitrov et al. (1999), Pseudobacciger presently is represented by just three

species: Pseudobacciger cablei Madhavi, 1975, P. manteri Nahhas & Cable, 1964 and P.

harengulae (Yamaguti, 1938). The present form is distinguished from two of these species as

follows. Pseudobacciger cablei is reported as a distinctly larger worm (528704 μm long vs

371453 μm for the present material) in which the suckers (especially the oral sucker) are actually

52

smaller than in the present form (Madhavi 1975). Pseudobacciger cablei differs further from the

new species in having intestinal caeca that terminate close to the anterior margins of the testes

rather than towards the posterior margin, and in having vitelline follicles that reach relatively close

to the anterior end of the body. Finally, P. cablei differs from the present form in being reported as

having the genital pore well anterior to the ventral sucker as opposed to immediately anterior to it.

Pseudobacciger manteri is reported at about the same overall size as the present material (373747

μmlong vs 371453 μm for the present material), but possesses a distinctly smaller oral sucker and

caeca that terminate well posterior to the testes (Nahhas and Cable 1964).

The present material is most similar to the type-species of Pseudobacciger, P. harengulae,

as originally described by Yamaguti (1938) from the clupeid Sardinella zunasi (Bleeker) (as

Harengula zunasi), and reported several times subsequently as reviewed by Dimitrov et al. (1999)

[We are doubtful that the form reported from a pomacanthid by Machida and Uchida (2001) as P.

harengulae can be conspecific with the species reported originally by Yamaguti (1938); see further

comments below]. The overall size and shape of P. harengulae as reported in those studies and the

present form are comparable as are the relative arrangements of their digestive systems and

reproductive organs, but we detect a difference, however, in the larger size of the suckers of the

present form: both oral and ventral suckers have size ranges either completely non-overlapping with

those reported for P. harengulae (see Gaevskaja 1996; Dimitrov et al. 1999) or overlapping with a

higher range (Yamaguti 1938). In particular, the ventral sucker of the present form has a length of

80109 μm, in contrast to 5885 μm in diameter as reported by Yamaguti (1938). Yamaguti’s

specimens appear to have been slightly flattened and it might thus be expected that the size of the

suckers would be inflated whereas they are distinctly smaller than reported here. Machida and

Uchida (2001) identified specimens from the pomacanthid Genicanthus lamarck Lacépède as P.

harengulae. These are much larger (9201,270 670890 μm) than previously described for this

species (and indeed for the genus), with proportionally larger dimensions for most characters,

including suckers, pharynx, oesophagus, gonads and eggs. In addition the forebody is relatively

longer than that of the new species (3545 vs 2334% of body length). Machida and Uchida (2001)

noted that the ‘variation with P. harengulae [as extended by their description] seems to make the

status of each of the three species rather obscure’. We think it is unlikely that Machida and Uchida’s

(2001) worms are conspecific with the tiny worm reported earlier from a clupeiform by Yamaguti

(1938)

Our conclusions regarding the identity of the present form are influenced by our

understanding of host-specificity of these parasites. Miller et al. (2011a) analysed the reported host-

specificity of 290 trematodes known from fishes of the Great Barrier Reef. Of these, just four had

been reported from more than one order of fishes and for none of the four had the wide specificity

53

been tested by molecular data. The high prevalence and intensity reported in the present study

suggest strongly that the infection of Chromis weberi is in no way accidental yet P. cheneyae n. sp.

has not been seen in a wide range of other pomacentrid species examined on the GBR. The three

presently-recognised species of Pseudobacciger are reported almost entirely from clupeiform

fishes; the exceptions are reports of P. harengulae from a herbivorous sparid (Perciformes, the

Bogue Boops boops Linnaeus) by Parukhin (1976a) and from a planktivorous pomacanthid

(Perciformes, Lamarck’s Angelfish Genicanthus lamarck) by Machida and Uchida (2001). For the

present form to occur widely in clupeiforms but in just one of many examined species of

Pomacentridae (Perciformes) seems highly anomalous. We thus think that the apparent distinctions

between the host of the present species and those reported previously for species of Pseudobacciger

are a likely indication that the present form is specifically distinct.

We consider taxonomic decision making in this genus to be unusually difficult. Species of

Pseudobacciger are usually exceptionally small (of the more than 300 species of trematodes

reported from fishes of the GBR, only the bivesiculid Paucivitellosus fragilis Coil, Reid & Kuntz,

1965 is as small as the new species (Pearson 1968), and important morphological features tend to be

obscured by eggs. We note that the analysis of Dimitrov et al. (1999) implied that P. harengulae is

an essentially cosmopolitan species given that there are records from Japan, Korea, the Atlantic and

Indian Oceans and the Black Sea. Although this is certainly possible, as demonstrated by the

molecular evidence that the aporocotylid Cardicola forsteri Cribb, Daintith & Munday, 2000 is

cosmopolitan (Aiken et al. 2007), such widespread species appear to be exceptional. We think it

possible, perhaps likely, that Pseudobacciger will prove to comprise a complex of significantly

more species than are presently recognised. Such a conclusion will almost certainly depend on the

generation of molecular data which, unfortunately, are available only for the present species so far.

In the light of analysis of the morphology and host-specificity of the present form we conclude that

it is best treated as a new species. However, the new name is proposed with a heightened sense that

all such proposals are hypotheses open to testing.

Systematics

The classification of the present form raises a systematic problem in that although its morphology

makes it convincing as a species of Pseudobacciger, it has been shown to be phylogenetically

remote from the representatives of four faustulid genera for which 28S or ITS2 rDNA sequence

data are available. Instead, its position, as can best be judged on the basis of molecular data, is

robustly within the Gymnophalloidea and as sister taxon to species of the Tandanicolidae. The

Tandanicolidae is a small family with species distinctive especially in the form of the terminal

genitalia and in usually infecting both marine and freshwater siluriforms (catfish) (Cribb and Bray

54

1994; Bray 2001). The present form certainly cannot be readily recognised in that family. The

gymnophallid metacercaria sequenced for this study formed the sister taxon to Pseudobacciger +

Tandanicolidae [We note that its identity as a gymnophallid was confirmed by its nesting among

several other identified gymnophallids in the analyses of ITS2 rDNA sequence data].

Gymnophallids are essentially parasites of birds (Scholz 2002) so that, although the present form is

very small, like most gymnophallids, it would have been a surprise if, of all the gymnophalloids, P.

cheneyae had proven most closely related to that family. Our analyses incorporated six genera of

Fellodistomidae but these form a well-supported clade sister to the Gymnophallidae +

(Pseudobacciger + Tandanicolidae). In addition to the Fellodistomidae, Gymnophallidae and

Tandanicolidae, Bray (2002) included the Botulisaccidae Yamaguti, 1971 and the Callodistomidae

Odhner, 1910 in the Gymnophalloidea; the latter two were included “for convenience of

identification”. No molecular data are available for the former family, but the sole member of this

monotypical family does not resemble Pseudobacciger cheneyae n. sp. The Callodistomidae, as

represented by the species Prosthenhystera obesa (Diesing, 1850) was found, utilising 28S rDNA,

to be the sister group to the Allocreadiidae Looss, 1902 within the superfamily Gorgoderoidea

Looss, 1901 by Curran et al. (2006). Phylogenetically, therefore, P. cheneyae n. sp. can thus be best

considered as Gymnophalloidea incertae sedis with the possibility that it may ultimately require a

family separate from those presently recognised. We speculate, thus, that the Faustulidae as

presently conceived may require division into two families, one related to the Zoogonidae as

presently recognised from the analysis of Olson et al. (2003) and a separate family within the

Gymnophalloidea.

Potential division of the Faustulidae has implications for the understanding of morphology

in the group. We observe that there is a potentially relevant dichotomy in the anatomy of the genera

presently recognised in the Faustulidae. Species of Allofellodistomum Yamaguti, 1971,

Pronoprymna Poche, 1926, Pseudobacciger and Pseudosellacotyle Yamaguti, 1953 have cirrus-

sacs or naked terminal genitalia with a very small pars prostatica whereas species of Antorchis,

Bacciger (but we note some variation within this genus), Echinobreviceca Dronen, Blend &

McEachran, 1994, Faustula, Paradiscogaster, Parayamagutia Machida, 1971, Trigonocryptus

Martin, 1958 and Yamagutia Srivastava, 1937 all have a cirrus-sac enclosing a large, prominent

pars prostatica. The three faustulid genera for which 28S rDNA sequences were previously

available (Antorchis, Bacciger and Trigonocryptus) and the single genus for which ITS2 rDNA

sequences were available (Paradiscogaster) are all of the latter type whereas P. cheneyae n. sp. is

of the former type. This difference may form a morphological basis for the recognition of separate

families. Notably, there are already two family group names, Pseudosellacotylinae Yamaguti, 1958

and Pentagramminae Yamaguti, 1958, which might be raised and applied to a distinct family within

55

the Gymnophalloidea. Although we consider the evidence too preliminary to justify a formal

proposal relating to family names at this stage, we feel unable to continue to recognise

Pseudobacciger in the Faustulidae. We recognise that the morphological similarity between P.

cheneyae n. sp. and other Pseudobacciger spp. may be convergent, but the similarity is so striking

that this seems unlikely.

Potential division of the Faustulidae also has implications for the understanding of life-

cycles in the group. A close relationship of the Faustulidae with the Zoogonidae has been

previously intriguing in that all the many known zoogonid life-cycles involve gastropods as first

intermediate hosts as recently reviewed by Barnett et al. (2014) whereas the handful of known

faustulid cycles (species of Bacciger and Pseudobacciger) involve bivalves as first intermediate

hosts (see Kim and Chun 1984; Bartoli and Gibson 2007). Should these species of Bacciger and

Pseudobacciger prove to form a clade within the Gymnophalloidea then that relationship would be

consistent with the use of bivalves by all known members of that clade, as well as the sister taxon

Bucephalidae. Such an outcome would also lead to the prediction that the Faustulidae that are

related to the Zoogonidae may prove to have gastropod rather than bivalve first intermediate hosts.

Ecology of damselfish and their parasites

The Pomacentridae is one of the most prominent families of fishes on the GBR with 79 species

known from the region (Randall et al. 1997). Pomacentrids are found mainly on shallow coral reefs,

but a few species have been recorded at depths of 80 metres or more. Despite this family being

diverse, occupying a range of habitats (live and dead coral, sand, coral rubble and the water-

column) (Wilson et al. 2008), having varying dietary preferences (herbivorous, omnivorous, and

planktivorous) (Randall et al. 1997), and a variety of social habits (social or solitary) (Fishelson

1998), the trematode fauna of this family is considered depauperate relative to some other fish

families (Barker et al. 1994). Although, pomacentrids have been extensively examined for

trematode parasites on the Great Barrier Reef (Manter 1969b; Cribb et al. 1992; Bray et al. 1993a;

Bray et al. 1993b; Bray and Cribb 1998; Bray and Cribb 2004; Hunter and Cribb 2012), no

gymnophalloid taxa have been recorded previously.

Trematodes of pomacentrids in Australian waters demonstrate a wide spectrum of patterns

of host specificity. Several species are evidently euryxenic (infecting taxa related only by ecology).

In this category are Hysterolecitha nahaensis Yamaguti, 1942, Lecithocladium moretonense Bray &

Cribb, 2004, Lepotrema clavatum Ozaki, 1932 and Thulinia microrchis (Yamaguti 1934) which

have all been reported from non-pomacentrids in addition to pomacentrids (Bray et al. 1993a;

Machida and Uchida 2001; Bray and Cribb 2002,2004) although the host range of these species

have not been tested by molecular approaches. Several species are stenoxenic (infecting

56

phylogenetically related host species); the haplosplanchnid Schickhobalotrema pomacentri (Manter,

1937) and the transversotrematid Transversotrema damsella Hunter & Cribb, 2012 have been

reported from multiple pomacentrid species (Cribb et al., 1994; Hunter & Cribb, 2012). Finally, two

species of Lepotrema Ozaki, 1932 (Lepocreadiidae), one species of zoogonid, and one species of

bivesiculid have each been reported from only a single pomacentrid species (Barker et al. 1994;

Cribb et al. 1994a; Bray and Cribb 1998). The specificity of P. cheneyae n. sp. is consistent with

this latter pattern of strict specificity. The ecology and behaviour of C. weberi is similar to that of

other planktivorous damselfish species, such as Abudefduf whitleyi Allen & Robertson, Chromis

ternatensis Bleeker and C. viridis Cuvier, all of which share the habitat of infected C. weberi. It is

thus unclear why P. cheneyae n. sp. should infect only C. weberi. As C. weberi is classified as a

non-territorial fish species (Ormond et al. 1996), it is unlikely that infection is the direct result of a

narrow habitat choice. Host-specificity is often thought of as being generated by encounter and

compatibility filters (Euzet and Combes 1980). In the present case it is not yet possible to determine

which filter might be driving the apparently strict specificity seen here.

Acknowledgements

We thank K. L. Cheney and E. C. McClure for their help collecting specimens and the staff of the

Lizard Island Research Station for their support. This work was funded by an Australian Research

Council Discovery grant awarded to T. H. Cribb and a Sea World Research and Rescue Foundation,

Australia (SWRRFI) awarded to D. Sun and T. H. Cribb.

57

Chapter 3: Presence of cleaner wrasse increases the

recruitment of fishes to coral reefs

58

Chapter 3: Presence of cleaner wrasse increases the recruitment of fishes to

coral reefs

Derek Sun, Karen L. Cheney, Johanna Werminghausen, Mark G. Meekan, Mark I. McCormick,

Thomas H. Cribb & Alexandra S. Grutter

Submitted to a journal for review

Abstract

Mutualisms affect the biodiversity, distribution and abundance of biological communities.

However, ecological processes that drive mutualism-related shifts in population structure are often

unclear and must be examined to elucidate how complex, multi-species mutualistic networks are

formed and structured. In this study, we investigated how the presence of key marine mutualistic

partners can drive the organisation of local communities on coral reefs. The cleaner wrasse,

Labroides dimidiatus, removes ectoparasites and reduces stress hormones for multiple reef fish

species, and their presence on coral reefs increases fish abundance and diversity. Such changes in

population structure could be driven by increased recruitment of larval fish at settlement, or by post-

settlement processes such as modified levels of migration or predation. We conducted a controlled

field experiment to examine the effect of cleaners on recruitment processes of a common group of

reef fishes, and showed that small patch reefs (61–285 m2) with cleaner wrasse had higher

abundances of damselfish recruits, than reefs from which cleaner wrasse had been removed over a

12-year period. However, the presence of cleaner wrasse did not affect species diversity of

damselfish recruits. Our study provides evidence of the ecological processes that underpin changes

in local population structure in the presence of a key mutualistic partner.

Introduction

On coral reefs, cleaning symbioses are key mutualistic networks that increase abundance and

promote species diversity in local fish communities (Grutter et al. 2003; Waldie et al. 2011).

Cleaner organisms, such as the bluestreak cleaner wrasse, Labroides dimidiatus, interact with

multiple species of fish, and provide fitness benefits to their clients by reducing stress hormones

59

(Soares et al. 2011) and ectoparasite loads such as gnathiid isopods (Grutter 1999a). The presence

of cleaner wrasse also affects fish population structure, size-frequency distribution and growth

(Grutter 1999a; Clague et al. 2011; Waldie et al. 2011). Abundance and diversity of fishes, both

adult damselfishes (site-attached) (Bshary 2003; Waldie et al. 2011) and visitors (moving between

patch reefs) (Grutter et al. 2003) increase, as does the abundance of juvenile visitor fishes (Waldie

et al. 2011) on reefs with cleaner wrasse.

Damselfishes (Pomacentridae) are one of the most abundant groups of fishes found on coral

reefs. Damselfishes play a critical role in structuring coral reef benthic communities (Ceccarelli et

al. 2005), with their presence indirectly increasing the survival of juvenile fishes (McCormick and

Meekan 2007). Although the abundance and diversity of adult damselfishes are higher on reefs

where cleaner wrasse are present (Bshary 2003; Waldie et al. 2011), it is not known if this

difference occurs during recruitment, as damselfishes are relatively sedentary after settlement (when

fish first join the reef assemblage after a pelagic larval stage). Whether differences in reef fish

population structure in the presence of cleaning organisms involve regulatory processes such as

migration or predation during settlement, or post-settlement during the juvenile and adult life

stages, has not been investigated. Understanding recruitment, a critical demographic process, is

important as it determines the structure of many marine and terrestrial communities (Caley et al.

1996). In this study, we examined whether the presence of cleaner wrasses influences recruitment

patterns of larval coral reef fishes on patch reefs, which could explain differences in the abundance

and diversity of local assemblages of adult fishes. As previous studies have shown a decrease in

abundance and diversity of reef fishes such as damselfishes and juvenile visitor fishes on reefs

where cleaner wrasse are absent (Bshary 2003; Grutter et al. 2003; Waldie et al. 2011), we predict a

similar trend occurring for damselfish recruits.

Material and methods

Our study was conducted on 16 small patch reefs in shallow (3−7 m depth) areas in lagoon and

back-reef habitats near Lizard Island, GBR, Australia. Since 2000, ‘removal’ reefs (n = 7) were

inspected at 3-month intervals and cleaner wrasse removed with hand and barrier nets; ‘control’

reefs (n = 9) were left undisturbed (see Clague et al. 2011 for a description of reefs; Waldie et al.

2011). The number of cleaner wrasse present on control reefs were recorded.

To investigate whether the abundance and species richness of newly-settled damselfishes

differed between removal and control reefs, we counted young recruits (< 20 mm TL) for five days

in each of three lunar cycles, beginning four days after the new moon (18−22 November 2012,

17−21 December 2012 and 16−20 January 2013). These times were selected because settlement of

60

fishes to coral reefs peak around the time of the new moon during these summer months (Milicich

and Doherty 1994). Counts alternated between randomly-selected control and removal reefs and

fish were identified to species where possible. Fish were counted once per reef by the same

observer (D.S.) on SCUBA for 60−120 min, depending on reef size. Fish were counted by the same

observer to reduce variability. As damselfishes are most active from morning to midday (Bosiger

and McCormick 2014), reefs were surveyed between 0900−1200 hr. The observer circled ~ 1 m

above the reef counting an abundant, or several less-abundant, species (in the same order for each

reef) on each pass. For analysis, fish were categorised either as common, wherein recruits of these

species were found on 90% of reefs across all three months (of which there were five species:

Pomacentrus adelus, P. amboinensis, P. moluccensis, P. nagasakiensis and Chrysiptera rollandi),

or uncommon (all other damselfishes). We compared the abundance and three measures of diversity

of damselfish recruits on control and removal reefs.

We conducted generalized linear mixed effect models (GLMER) with a Poisson distribution

error, to examine whether the total abundance of all the five common damselfishes combined (and

also separately by species) and the total abundance of uncommon damselfishes differed relative to

the presence of cleaner wrasse. The terms treatment (cleaner or no cleaner presence), site (Lagoon

or Casuarina Beach) and time (November 2012, December 2012 and January 2013) were fixed

factors; reef identity was a random factor and reef area was a covariate. A posteriori Tukey-Kramer

HSD post-hoc test was used to interpret significant interactions. We also used full linear mixed-

effect models (LMER) with a REML procedure to analyse species richness and Simpson’s indices

of diversity (Krebs 1999) and used a permutational multivariate analysis of variance

(PERMANOVA) and principle coordinates analysis (PCO) for the species composition. To

determine the best model in all analyses, we compared the full model with models in which one of

the explanatory terms was dropped using the “drop1” function (Chambers 1992). The term was

dropped if the analysis of variance found that a dropped term had no significant effect on the model

using the Chi-square distribution. A Bonferroni correction (α = 0.01) was applied to account for

multiple comparisons. All statistical analyses were conducted using R version 3.0.1 (R

Development Core Team 2013), except for the species composition which was analysed using

PRIMER 6.0 (Anderson et al. 2008).

Results

Thirty-one species of damselfish recruits were identified (see Appendix I). The total abundance of

common damselfishes was lower on removal than on control reefs (p = 0.006, Figure 1). At the

species level, the abundances of C. rollandi (p = 0.0001, December only) and P. amboinensis (p <

61

0.0001, Figure 2a-b) were significantly lower on reefs where cleaner wrasse had been removed than

on control reefs. Although there were similar trends in the average abundance of the other common

species, P. adelus, P. moluccensis and P. nagasakiensis, these differences were not significant (p =

0.0769, 0.0689 and 0.0573 respectively, Figure 2c-e). Total abundance of uncommon damselfishes

was also lower on removal reefs (p = 0.0070, Figure 2f).

We found no differences in measures of diversity between removal and control reefs (species

richness: p = 0.3578; Simpson’s diversity index: p = 0.8418; species composition: PERMANOVA,

F (1 ̶36) = 1.83, p = 0.1029, Figure 3a-c).

Figure 1 Total abundance of common damselfishes on experimental reefs. Means (± SE)

abundances are from reefs with and without cleaner wrasse. * Significant difference in cleaner

wrasse treatment (p < 0.01).

62

Figure 2 Mean (± SE) abundance of recruiting damselfish species. Control and cleaner wrasse

removal reefs: a-e) abundance of individual common species, f) abundance of uncommon species. *

Significant difference in cleaner wrasse treatment (p < 0.01), n.s = p > 0.01.

63

Figure 3 Diversity of all damselfish recruits per reef. a) Species richness, b) Simpson’s diversity

index, c) species composition on control and cleaner wrasse removal reef, represented by a principle

coordinate analysis (PCO).

64

Discussion

The combined abundances of the five common species, the individual abundances of two common

species (C. rollandi and P. amboinensis), and the total abundance of uncommon recruited

damselfishes were higher on reefs with cleaner wrasse present than on those without. There was

also a strong (but not statistically significant) trend for abundances of the other common

damselfishes (P. adelus, P. moluccensis and P. nagasakiensis) to be higher on reefs with cleaners.

These results suggest that the presence of a key mutualistic species, creates changes in population

structure around the time of settlement and recruitment. This in turn could explain the observed

changes in abundance and distribution of coral reef fish populations in relation to cleaner wrasse

presence (Waldie et al. 2011).

The presence of cleaner wrasse could influence recruitment processes by a number of direct

or indirect mechanisms. If young fish use visual or olfactory cues to locate potential microhabitats,

then sensory cues may enable them to locate cleaners directly. Indeed, a recent study suggests

recently-settled fish can recognise and preferentially choose a microhabitat near a cleaner (Sun et al.

in review, Chapter 4). Alternatively, cleaner wrasse could influence recruitment processes indirectly

by reducing the abundance of ectoparasites in the local environment. Higher abundances of

parasites in a system have the potential to detrimentally affect the recruitment of juveniles, as

demonstrated in lower recruitment success of juvenile cricket frogs (Acris crepitans) in ponds with

high rates of parasitic infection (Beasley et al. 2005). Although both adults and recruit damselfishes

are known to be infected with gnathiids, cleaner wrasse rarely clean recruits and tend to clean larger

individuals (Sun et al. in review, Chapter 4). However, with gnathiids decreasing swimming

performance and increasing mortality of settling fish (Grutter et al. 2011) the removal of even a

single gnathiid can increase the survival of an individual. Cleaner wrasse may also indirectly

promote fish recruitment by influencing the abundance of conspecifics, which have been shown to

influence microhabitat selection by settling fish, as settling fish may use the presence of

conspecifics as a measure of habitat quality (Coppock et al. 2013) or the behaviour of predators

(Cheney et al. 2008). The abundance of all (adult and juvenile) conspecifics for P. amboinensis, P.

moluccensis and P. nagasakiensis are also higher on reefs with cleaner wrasse than those without

(D. Sun, unpublished data). Further investigation is required to disentangle the relative importance

of these direct and indirect mechanisms. It should be noted that the gradual decrease in the

abundance of damselfish recruits from November to January is due to the larval fish season ending.

Interestingly, there was no difference in species richness, Simpson’s diversity index, or

composition of damselfish recruits between control and removal reefs, suggesting that the presence

of cleaner wrasse affects the abundance of individual species but not the overall community

diversity of recruit damselfishes. Previous studies have shown an increase in the abundance and

65

species richness of resident adult fishes in the presence of cleaner wrasse (Bshary 2003; Waldie et

al. 2011). Our results suggest that such patterns in diversity are not likely due to differential

recruitment, but due to post-settlement events, including migration in response to cleaner wrasse

presence or the effects of parasitism (Waldie et al. 2011). Parasitic infection by gnathiid isopods can

make recruits more susceptible to the high rates of predation that occur immediately after settlement

(typically 56 % within two days of settlement) (Almany and Webster 2006). Ultimately, this

process could affect the diversity of fishes in the presence of cleaner wrasse.

In summary, we have demonstrated for the first time that the presence of cleaner wrasse has

the potential to influence the structure and demography of reef fish populations via effects on the

process of recruitment. Thus, mutualism is an important mechanism that contributes to the already-

complex recruitment processes of reef fishes.

Acknowledgements

We thank Lizard Island Research Station staff. We also like to thank E. C. McClure, M. De

Brauwer for field assistance. S.P. Blomberg for statistical advice and R. Q-Y. Yong for reading

drafts of this manuscript. Funding was provided by Sea World Research and Rescue Foundation,

Australia (SWRRFI) to D.S, T.H.C and A.S.G and an Australian Research Council Discovery Grant

to A.S.G and M.G.M.

66

Chapter 4: Cleaner wrasse influence habitat selection of

young damselfish

67


Derek Sun, Kare L. Cheney, Johanna Werminghausen, Eva C. McClure, Mark G. Meekan, Mark I.

McCormick, Thomas H. Cribb & Alexandra S. Grutter

Submitted to a journal for review

Abstract

The presence of bluestreak cleaner wrasse, Labroides dimidiatus, on coral reefs increases total

abundance and biodiversity of reef fishes. The mechanism(s) that cause such shifts in population

structure are unclear, but it is possible that young fish preferentially settle into microhabitats where

cleaner wrasse are present. To investigate this possibility, we conducted aquarium experiments to

examine whether settlement-stage and young juveniles of ambon damselfish, Pomacentrus

amboinensis, selected a microhabitat near a cleaner wrasse (adult or juvenile). Both settlement-stage

(0 d post-settlement) and juvenile (~ 5 weeks post-settlement) fish spent a greater proportion of

time in a microhabitat adjacent to L. dimidiatus than in one next to a control fish (a non-cleaner

wrasse, Halichoeres melanurus) or one where no fish was present. This suggests that cleaner wrasse

may serve as a positive cue during microhabitat selection. We also conducted focal observations of

cleaner wrasse and counts of nearby damselfishes (1 m radius) to examine whether newly-settled

fish obtained direct benefits, in the form of cleaning services, from being near a cleaner wrasse in

the field. Although abundant, newly-settled recruits (< 20 mm TL) were rarely (2 %) observed

being cleaned per 20 min observations compared with larger damselfishes (58 %). Individual

damselfish that were cleaned were significantly larger than the median size of the surrounding

nearby non-cleaned conspecifics; this was consistent across four species. The selection by

settlement-stage fish of a microhabitat adjacent to cleaner wrasse in the laboratory, despite only

being rarely cleaned, suggests that even rare cleaning events and/or indirect benefits may drive their

settlement choices. This behaviour may also explain the decreased abundance of young fishes on

reefs from where cleaner wrasse had been experimentally removed. This study reinforces the

potentially important role of mutualism during the processes of settlement and recruitment of young

reef fishes.

68

Introduction

On coral reefs there are many mutualistic cleaners that remove parasites from other organisms,

including coral reef fishes. One of the most obvious of these is the bluestreak cleaner wrasse,

Labroides dimidiatus (Labridae), a cleaner wrasse common throughout the Indo-Pacific (Randall et

al. 1997). Labroides dimidiatus has been shown to reduce the ectoparasite load of individual client

fishes (Gorlick et al. 1987; Grutter 1999a; Clague et al. 2011), aggression by predators (Cheney et

al. 2008), stress (measured in cortisol levels) via physical contact (Bshary et al. 2007; Soares et al.

2011) and increase growth and size of client fishes (Clague et al. 2011; Waldie et al. 2011). In

addition, these cleaner wrasse affect local populations of reef fishes, so that in their absence there

are reductions in the diversity and abundance of site-attached adult residents, visitors (fishes that

move between patch reefs) and juvenile visitor fishes (Bshary 2003; Grutter et al. 2003; Waldie et

al. 2011). Although the positive effects of the presence of L. dimidiatus on individuals and

populations have been demonstrated in experimental manipulations of patch reefs ranging in size

from ~61 to 285 m2, the mechanisms involved, particularly at small spatial scales (1 m–10s of m),

remain largely unstudied (Waldie et al. 2011).

One of the most critical periods that effects the population dynamics of coral reef fishes

occurs during the settlement of young fish from the plankton to the reef and the few days or weeks

of life in benthic habitats that immediately follow this event (post-settlement). Processes at this time

ultimately determine which individuals are recruited to the population of juveniles. Young fish can

actively choose the habitat into which they will settle and the processes that underlie these choices

are complex (Hoey and McCormick 2004; Heinlein et al. 2010). Additionally, the mortality rate at

this time is high; indeed, it has been estimated that 57 % of individuals die within the first 1−2 days

of settlement (Almany and Webster 2006), which could be due to microhabitat selection. Thus, the

choice of a suitable microhabitat at settlement is critical for successful recruitment. Selecting the

optimum microhabitat depends on numerous factors, which may include the structure of the

microhabitat (Tolimieri 1995), and the presence of potential predators (Vail and McCormick 2011)

and conspecifics (Sweatman 1985). It was recently shown that the experimental removal of L.

dimidiatus was associated with a decrease in the abundance of new recruits on patch reefs, relative

to control reefs where cleaner wrasse were present (Sun et al., in review, Chapter 3). However, it is

unknown whether the presence of L. dimidiatus can serve as a positive cue for microhabitat choice

at settlement.

Our study used experimental and observational approaches to determine whether the

presence of juvenile or adult cleaner wrasse influenced the choice of microhabitat by settling

juveniles of the ambon damselfish, Pomacentrus amboinensis. In the laboratory, we examined

whether the presence of cleaner wrasse influences habitat choice of P. amboinensis. To determine

69

the extent to which recently-settled damselfishes were cleaned by L. dimidiatus juveniles, and thus

whether the benefits of settling near a cleaner wrasse could be related to the cleaning services they

provide or to other indirect benefits, focal observations of the cleaning interactions between L.

dimidiatus and damselfishes were conducted in the field. Lastly, to determine whether juvenile

cleaner wrasse are size selective in their choice of clients, we compared the size of individuals of

the four most abundant damselfishes that were cleaned by the wrasse with the median size of non-

cleaned conspecifics.


Laboratory experiment: Does the presence of L. dimidiatus affect microhabitat choice in young

damselfish?

We conducted an aquarium experiment at Lizard Island Research Station, GBR, Australia, to

investigate whether the presence of L. dimidiatus influenced microhabitat choice of P. amboinensis,

a species chosen because of its high abundance at the study site (Sun et al. in review, Chapter 3).

Pomacentrus amboinensis were collected between November 2011 and January 2012 using light-

traps (see Meekan et al. 2001 for design) moored overnight approximately 100 m from the reef.

These fish were juveniles caught just prior to settlement from the plankton and thus had no previous

exposure to L. dimidiatus. Fish were held in groups of ~ 20 in clear plastic holding aquaria (29 × 17

× 12 cm) with a constant flow of seawater. Fish collected in November were held for around 5

weeks (at which point they were classified as juveniles, i.e. ~ 5 weeks post-settlement), whereas

fish collected in January were tested within a day of capture (classified as settlement-stage, i.e. 0

days post-settlement). Fish were fed ad libitum twice a day with live nauplii of brine shrimp

(Artemia sp.).

Both adult and juvenile cleaner wrasse (L. dimidiatus) and a non-cleaner control fish

(pinstripe wrasse, Halichoeres melanurus), were collected from the Lizard Island lagoon using

hand-nets, barrier nets and anaesthetic clove oil mixed with alcohol and seawater (10 % clove oil;

40 % ethanol; 50 % seawater). Halichoeres melanurus was chosen as a control due to similarity in

body shape to L. dimidiatus (Grutter 2004a; Cheney et al. 2008; Cheney et al. 2009). All wrasses

were maintained in aquaria (43 × 32 × 30 cm) with running unfiltered seawater.

Clear Perspex sheets were placed at either end of a glass experimental aquaria (35 × 36 × 65

cm), with flow-through seawater, to create two end compartments (10 cm away from the end of the

aquarium). The middle section of the aquarium was divided into three 15 cm subsections, using

vertical black lines drawn on the front and back of the aquarium. Three polyvinyl chloride (PVC)

tube shelters (6 × 5 cm diameter) were placed in the centre of each subsection (Fig. 1). Aquaria

70

were covered with black plastic on three sides to isolate the fish from external activity, and a green

plastic shade cloth hung in front of the tanks to create an observation blind.

In each end compartment a L. dimidiatus, a control fish (H. melanurus) or no fish was

placed, depending on the treatment: 1) L. dimidiatus and H. melanurus; 2) L. dimidiatus and no

fish; or 3) H. melanurus and no fish. These combinations were presented to a randomly-selected P.

amboinensis ‘client’ and the side in which a cleaner or control (or no fish) was placed was

randomised, but balanced. The ‘no fish’ treatment at one end compartment of the aquarium was

used to control for the ‘client’ potentially avoiding any heterospecifics, irrespective of identity.

A P. amboinensis was placed in the middle section of the aquarium in a bottomless clear

plastic container (8 × 8 × 8 cm) with numerous small (5 mm) ventilation holes for 20 min prior to

commencing the trial, allowing the fish to acclimate to the new surroundings. The bottomless

container was removed using an attached monofilament string by the observer, who was positioned

behind the observation blind. Each trial was conducted for 20 min, and the position of P.

amboinensis within the aquarium (left, middle, or right section) was recorded every 15 s. It is

inferred that the degree of proximity to the stimulus represented its choice of microhabitat.

Experiment 1 was divided into three parts (A, B, C) with different ontogenetic stages of

cleaner, control and P. amboinensis per treatment (see Table 1 for a summary of the fish stage and

species). For each part, we used 8 cleaner and 8 control fish that were randomly selected, and 75 P.

amboinensis (n = 25 individuals per treatment). At the end of each trial, cleaner and control fish

were replaced. Individual P. amboinensis were only used once, so that their behaviour was not

affected by previous exposure.

Part A used adult L. dimidiatus and H. melanurus, and Part B used juveniles of both species.

These experiments tested whether the ontogenetic stage/size of L. dimidiatus affected habitat choice

of juvenile P. amboinensis. Part C was conducted when settlement-stage fish became available in

the light-traps; juvenile L. dimidiatus and H. melanurus were used for this component. The aim of

this experiment, relative to Part B, was to determine whether P. amboinensis’ ontogenetic stage/size

affected its habitat choice.

71

Figure 1 Front view of the laboratory experiment tank design. A: clear Perspex sheet placed on

either side to create compartments for stimuli, B: vertical dashed lines on front and back of

aquarium to create subsections, and C: PVC tube shelters placed in each subsection. Not drawn to

scale.

Table 1 Ontogenetic stage and size of fish used in each part for laboratory experiment. Size is mean

± SE, TL.

Part Date Labroides

dimidiatus

Halichoeres

melanurus

Pomacentrus

amboinensis

(cleaner) (control) (client)

A December 2011 Adult

(79.1 ± 4.0 mm)

Adult

(86.1 ± 0.8 mm)

Juvenile

(19.3± 0.3 mm)

B December 2011 Juvenile

(24.2 ± 0.5 mm)

Juvenile

(24.8 ± 0.4 mm)

Juvenile

(18.0 ± 0.3 mm)

C January 2012 Juvenile

(23.1 ± 0.7 mm)

Juvenile

(24.3 ± 0.8 mm)

Settlement-stage

(15.2 ± 0.1 mm)

72

Field observations: Do juvenile L. dimidiatus clean recently-settled damselfish recruits and are

they selective in their choice of clients?

To assess whether recently-settled damselfish recruits are cleaned by L. dimidiatus, and thus gain

direct fitness benefits from settling near them, we conducted 20 min behavioural observations on

juvenile L. dimidiatus (n = 79) on a section of continuous reef (~ 200 m) at Coconut Beach

(1440S, 14528E), Lizard Island. Although cleaner wrasse of all sizes mainly consume

ectoparasites (Grutter 2000), we observed juvenile L. dimidiatus, rather than adults, due to the

higher frequency of juvenile L. dimidiatus interacting with young reef fish (Robertson 1974, D.

Sun, unpublished data). Before observations commenced, observers learned to accurately estimate

fish size (TL) underwater by using printed and laminated outlines of model fish (12−74 mm TL).

The fish models were placed on the reef at varying distances from the observer, and sizes were

estimated and compared with the actual length of the model until observers were more than 80 %

accurate at estimating length. Rulers printed onto dive slates were also used as a guide for

estimating fish length.

For each observation, a juvenile L. dimidiatus was located and its estimated TL was

recorded. A 1 m radius was measured around the area where the L. dimidiatus was first sighted

using a 1 m string and marked using five lead sinkers (7 mm) attached to floats (25 mm) and all fish

within this area were recorded. All fish present were recorded to account for the availability of other

sizes of fish as potential clients that could influence the cleaners’ choice of client. Surrounding fish

were allowed to acclimate (~ 5 min) to the presence of the markers and observer, before the

commencement of observations. Cleaning interactions were recorded as any physical contact

between cleaner and client, or significant inspection by the cleaner (> 1 s). For each interaction, we

recorded the client species (where possible), cleaning duration, and estimated client size (TL, mm).

To compare client size and the median size of nearby (non-cleaned) conspecifics, the size of all

other fish present within the marked observation site (TL = > 10 to < 80 mm), but not involved in

any cleaning interaction, was also estimated. All other fish were identified to species where

possible. To ensure that each L. dimidiatus was observed only once, observers began at one end of

the reef and continued without backtracking. In contrast to adults, juvenile L. dimidiatus stay within

a highly restricted home range of around 4 m2 (Robertson 1974), further reducing the possibility of

observing an individual twice. Although every individual within the observation area was identified

to species, ultimately, in order to adequately perform statistical analyses, only the most abundant

damselfish species (both cleaned and non-cleaned conspecifics) from the observations were used.

These four common species were Pomacentrus amboinensis, P. moluccensis, P. nagasakiensis and

P. wardi.

73

Statistical analyses

Laboratory experiment

A full linear mixed-effects (LMER) model with a restricted maximum likelihood (REML)

procedure was used to examine whether P. amboinensis spent more time near a L. dimidiatus than

near a control fish or no fish compartment. Two analyses were conducted; the first model used data

from parts A and B (Table 1) to determine whether ontogenetic stage of the cleaner had an effect on

juvenile P. amboinensis’ habitat choice, while the second model used data from part B and C, and

included ontogenetic stage (settlement-stage and juvenile P. amboinensis) of the client as a factor.

In the first model, cleaner ontogenetic stage (adult or juvenile), settlement stimulus type to which

the microhabitat was adjacent (cleaner, control, no fish), aquarium side the cleaner was placed (left

or right) and fish stimulus treatment combination (cleaner/control, cleaner/no fish or control/no

fish) were used as fixed factors. Replicate trial identity was added as a random factor, to account for

an order effect. In both models, proportion of time spent in each section of the aquarium (right, left

and centre) was transformed by taking the arcsine of the square root of the proportion to meet the

assumptions of normality.

Field observations

A LMER model with a REML procedure was used to test whether there was a significant difference

between client fish and the median size of the nearby conspecifics. The terms treatment (client or

nearby conspecifics), nearby conspecific size and species (four damselfish species: P. amboinensis,

P. moluccensis, P. nagasakiensis and P. wardi) were fixed factors, and the identity of the individual

cleaner fish and the cleaning event identity (specifying which client and nearby conspecifics

corresponded with one another) were random factors. To determine whether client size varied with

cleaner size or the size of all other nearby species, separate (LMER) analyses were run with cleaner

size and other nearby fish size as covariates and cleaner identity as a random factor. A general

linear model (GLM) with a binomial distribution was used to determine whether there was a

significant difference between recruits that were cleaned and the abundance of other fish species

and recruits within the 1 m radius observation area.

To determine the best model in all analyses, we compared the full model with models in

which one of the explanatory terms was dropped using the ‘drop1’ function (Chambers 1992). The

term was dropped if the analysis of variance found that a dropped term had no significant effect on

the model using the Chi-square distribution. A Tukey-Kramer HSD test (TK-HSD) using the glht

function in the ‘multcomp’ package (Horthorn et al. 2013) identified where differences occurred

and a summary output table was reconstructed using the results from the ‘drop1’ function. Prior to

all analyses, the assumptions of normality and homogeneity of variance were assessed using

74

histograms, residuals and quantile-quantile plots. All statistical analyses used R version 3.0.1 (R

Development Core Team 2013).

Results

Laboratory experiment: Does the presence of L. dimidiatus affect microhabitat choice in young

damselfish?

In all three parts (A-C) of the experiment, P. amboinensis spent significantly more time in the

subsection or PVC tube that is adjacent to a L. dimidiatus than next to a control fish or no fish

compartment (Fig. 2a-c). The first statistical analysis, which used data from parts A and B to

examine whether cleaner ontogenetic stage had an effect, showed that the proportion of time that

juvenile P. amboinensis spent next to the stimulus fish (cleaner, control or no fish) differed

according to an interaction between fish stimulus and cleaner stage (LRT = 19.98, df = 2, P <

0.0001; Fig. 2a, b); a TK-HSD test showed that juvenile P. amboinensis spent significantly more

time near a juvenile than an adult L. dimidiatus (P < 0.05). There was no significant effect of

treatment combination (LRT = 0.474, df = 2, P = 0.788) or aquarium side (LRT = - 4.18, df = 1, P =

0.772).

For the second statistical analysis, which used data from parts B and C to examine whether

ontogenetic stage of the client affected habitat choice, the proportion of time that P. amboinensis

spent next to a habitat did not differ with stage (LRT = - 4.40, df = 1, P = 0.749) but did differ

according to fish stimulus (LRT = 144.22, df = 2, P < 0.0001; Fig. 2a, c); a TK-HSD test showed

that P. amboinensis spent significantly more time near L. dimidiatus (P < 0.05). The proportion of

time spent in a habitat differed according to treatment combination (LRT = 37.11, df = 2, P <

0.0001). A TK-HSD test showed that P. amboinensis spent more time in both chosen microhabitats

(i.e. spent less time in the middle section) in the cleaner/no fish and control/no fish stimulus

treatment combinations than in the cleaner/control treatment. The effect of aquarium side was not

significant (LRT = -1.44, df = 1, P = 0.182). There was no significant effect of size between cleaner

and control wrasse (> 0.05).

75

Figure 2 The percentage of time spent on microhabitat adjacent to fish treatment by damselfish

client Pomacentrus amboinensis. a) Part A: Juvenile P. amboinensis with adult cleaner/control fish,

b) Part B: Juvenile P. amboinensis with juvenile cleaner/control fish, and c) Part C: Settlement-

stage P. amboinensis with juvenile cleaner/control treatment.

76

Field observation: Do juvenile L. dimidiatus clean recently-settled damselfish recruits and are

they selective in their choice of clients?

A total of 1107 cleaning interactions were recorded during 79 observations of individual L.

dimidiatus juveniles over a period of 1580 min. Cleaner wrasse cleaned 104 species of fish from 19

families, including 32 species of damselfishes of which the following were cleaned most often

(listed from most to fewest): Acanthochromis polyacanthus, Stegastes apicalis, Amblyglyphidodon

curacao, P. moluccensis, Plectroglyphidodon lacrymatus, P. amboinensis, P. nagasakiensis, P.

wardi and Chromis lepidolepis (see Appendix 2 for a full list of client species). These interaction

frequencies were not adjusted for relative abundance of each client species. The mean number (±

SE) of clients cleaned and the duration spent cleaning per 20 min observation was 14 ± 1 fish and

83.8 ± 7.3 s, respectively. Overall, only 14 % of cleaning interactions were between juvenile L.

dimidiatus and the four common damselfishes (P. amboinensis, P. moluccensis, P. nagasakiensis

and P. wardi) that had the greatest abundances within the 1 m radius of a cleaner wrasse. A total of

46 % of interactions occurred with other damselfishes (excluding the latter common species listed),

while the remaining 40 % of interactions occurred between non-damselfishes and juvenile L.

dimidiatus. Overall, 60 % of the clients of juvenile L. dimidiatus were damselfishes. The range and

median size (mm TL) of the common damselfishes that were cleaned were: P. amboinensis (20–80

and 50 mm), P. moluccensis (25–80 and 50 mm), P. nagasakiensis (15–90 and 40 mm), and P.

wardi (50–120 and 90 mm); the duration (seconds, mean ± SE) of time spent cleaning per 20 min

observation for the common species was 14 ± 2.5 s. Client size differed among the four species

(LRT = 68.01, df = 3, P < 0.0001; Fig. 3a); a TK-HSD test showed that P. wardi clients were

significantly larger than the other common client damselfishes. Damselfishes that were cleaned

were significantly larger than nearby non-cleaned conspecifics occurring within a 1 m radius; this

effect was consistent for all four species (LRT = 173.72, df = 1, P < 0.0001; Fig. 3a). Separate

analyses showed that the size of the damselfish client was not correlated with the size of the cleaner

wrasse (LRT = 0.17, df = 1, P = 0.675) nor was it correlated with the median size of all other fish

species combined (excluding common damselfishes) within the 1 m radius of the cleaner (LRT =

2.55, df = 1, P = 0.110).

Despite the presence of numerous recently-settled damselfishes (size range 15–20 mm)

within the 1 m radius observation areas (Fig. 3b), only 2 % (22 out of 1107 cleaning interactions) of

fishes that were cleaned were < 20 mm TL. There was a significant negative relationship between

the number of recruits that were cleaned and the abundance of both other fish species (> 20 mm,

TL) (LRT = 7.97, df = 1, P = 0.004, Fig. 4a) and recruits (LRT = 16.95, df = 1, P = < 0.0001, Fig,

4b) within the 1 m radius observation area. The probability of a recruit (< 20 mm) being cleaned by

a juvenile cleaner wrasse in a 20 min period is 0.002 (95 % CI = 3.22 × 10-6 ̶ 0.22) when there are

77

nine other fishes and a probability of 0.003 (95 % CI = 6.77 × 10-5 ̶ 0.07) when there are three

recruits within the observation area (Fig. 4a-b).

Figure 3 Cleaning interactions of juvenile cleaner wrasse, Labroides dimidiatus and damselfishes.

a) Size (TL) of damselfish clients and nearby non-cleaned conspecifics. Bars represent median;

error bars represent 25th and 75th quantiles, b) size distribution of nearby non-cleaned damselfish per

1 m radius; bars indicate mean, error bars represent SE.

<10

11-1

5

16-2

0

21-2

5

26-3

0

31-3

5

36-4

0

41-4

5

46-5

0

51-5

5

56-6

0

61-6

5

66-7

0

71-7

5

76-8

0>8

00

2

4

6

8

Size categories (mm)Mean

ab

un

dan

ce o

f d

am

self

ish

per

1m

rad

ius

b

P. a

mboin

ensi

s

P. m

olucc

ensi

s

P. n

agas

akie

nsis

P. war

di0

20

40

60

80

100

Fish species

Med

ian

siz

e (

mm

) Client

Nearby

a

78

Figure 4 The probability of an individual recruit (< 20 mm) being cleaned by a juvenile cleaner

wrasse, Labroides dimidiatus, in relation to the abundance of a) other fish species and b) abundance

of recruits per 20 min observations.

Discussion

Our study provides experimental evidence that the presence of cleaner wrasse, L. dimidiatus, may

directly influence the choice of microhabitats of young coral reef damselfish. In the laboratory, both

settlement-stage and juvenile P. amboinensis preferentially selected microhabitats that were in close

proximity to L. dimidiatus. This may explain the observed patterns of increased abundance of fishes

in the presence of cleaner wrasse on the reef. However, contrary to expectations, our observations

revealed that these young damselfish were rarely cleaned by L. dimidiatus. Thus, it is possible that

Contr

ol

Rem

oval

0

20

40

60

80

100

P. amboinensis

Me

an

ab

un

da

nc

e p

er

ree

f

1a*

Contr

ol

Rem

oval

0

20

40

60

80

100

P. moluccensis

Me

an

ab

un

da

nc

e p

er

ree

f

1cn.s.

Contr

ol

Rem

oval

0

5

10

15

20

P. adelus

Me

an

ab

un

da

nc

e p

er

ree

f

1en.s.

Contr

ol

Rem

oval

0

10

20

30

C. rollandi

Me

an

ab

un

da

nc

e p

er

ree

f

1b*

Contr

ol

Rem

oval

0

20

40

60

80

100

P. nagasakiensis

Me

an

ab

un

da

nc

e p

er

ree

f

1d*

Contr

ol

Rem

oval

0

20

40

60

80

100

Total damselfish abundance

Me

an

ab

un

da

nc

e p

er

ree

f

1f*

79

even these rare cleaning events and/or indirect benefits resulting from the presence cleaner wrasse

may drive the selective behaviour of settling damselfish.

We assessed microhabitat choice by settlement-stage (0 days post-settlement) and slightly

older juvenile (~5 weeks post-settlement) P. amboinensis and found that individuals preferentially

selected a microhabitat adjacent to L. dimidiatus. This indicates that the ‘attraction’ towards cleaner

wrasse occurs in both stages. Habitat selectivity at settlement appears common in young

damselfishes, with some species attracted to particular microhabitats such as coral heads that

harbour conspecifics (Lecchini et al. 2007a; Coker et al. 2012a), while others are attracted to coral

heads associated with heterospecifics (Green 1998; Almany 2003). Our study adds cleaner wrasse

to the list of organisms that settlers are attracted to, so that the presence of cleaner wrasse could act

as an indicator of microhabitat quality to settlement-stage fish. Furthermore, these results suggest

that if juveniles move between microhabitats, that their site selection may also be influenced by the

presence of L. dimidiatus.

The apparent attraction of L. dimidiatus may explain the greater number of recently-settled

recruits (Sun et al. in review, Chapter 3), juvenile visiting fishes and even adult damselfishes, found

on patch reefs where L. dimidiatus are present relative to reefs without cleaner wrasse (Bshary

2003; Waldie et al. 2011). There are two ways in which these patterns may arise, using the presence

of L. dimidiatus as a cue. Settlement-stage fish may either choose to settle on certain patch reefs, or

engage in post-settlement movement between patch reefs as juveniles (Simpson et al. 2008). Both

of these processes may additively result in damselfishes being more abundant on reefs with cleaner

wrasse than on those without.

The fish used in the laboratory experiment were naïve with respect to the shape or odour of

L. dimidiatus, since they were collected by light-traps while still in their planktonic phase. Despite

this naïveté, young fish were able to differentiate between L. dimidiatus and another species that

was similar in size and shape but differed in colour, pattern and behaviour. The attraction to cleaner

wrasse thus appears to be an innate behaviour. Slightly older recruits displayed the same abilities.

However, we did not test whether P. amboinensis recognised L. dimidiatus as a cleaner per se. On

some occasions, the fish did behave like a client by remaining in the same position and adopting a

pose with spread fins when next to the compartment with a cleaner wrasse. This behaviour did not

occur when they were adjacent to the control fish or an empty compartment. Prospective clients of

cleaner fish often assume this posture when they attempt to attract a cleaner to inspect them (Côté et

al. 1998). Because the cleaner wrasse was separated by Perspex, both the cleaner and client were

unable to physically interact; such physical contact likely acts as a positive feedback (Bshary and

Würth 2001). Losey et al. (1995) previously showed that newly-metamorphosed laboratory-reared,

and hence ‘cleaner-naïve’ Hawaiian humbug damselfish, Dascyllus albisella, recognised the

80

Hawaiian cleaner wrasse, Labroides phthirophagus by adopting the cleaning ‘pose’. Our study thus

confirms the hypothesis that client fish have an innate ability to recognise cleaners.

It seems likely that P. amboinensis used visual cues to differentiate between the two

stimulus fish that were presented. A combination of black and blue and/or yellow body patterns

displayed by cleaner fish, including L. dimidiatus, allows clients to visually recognise cleaners

(Cheney et al. 2009). Indeed, the particular colouration and patterns displayed by cleaner fish attract

the higher numbers of client fish than other colour and pattern combinations. Client fish are also

known to recognise cleaner fish based on their small body size and the presence of lateral stripes

(Stummer et al. 2004).

Pomacentrus amboinensis both chose adult and juvenile L. dimidiatus over the other stimuli.

This occurred despite the two cleaner stages being differently coloured, with both having a black

stripe contrasted by blue, but the adults also having the colours yellow and white. However,

juvenile P. amboinensis spent more time near juvenile L. dimidiatus than adults implying the

possibility of ontogenetic variation in choice of cleaner fish.

Alternatively, this could simply be due to the difference in relative size between the client

and cleaner at different ontogenetic stages. The same choice test was not performed using

settlement-stage fish due to their limited numbers in the light traps. Previous studies have shown

that fish larvae have good visual senses, which enables them to detect and recognise conspecifics,

heterospecifics and predator fishes (Lecchini et al. 2005; Lecchini et al. 2014).

Focal observations of juvenile L. dimidiatus on reefs revealed that only 2 % of their clients

were < 20 mm TL. This suggests that small individuals are rarely cleaned by L. dimidiatus and so

the interaction may contribute little to the cleaner wrasse’s diet. Small-bodied individuals, such as

recently-settled damselfishes, are infected with very few ectoparasites (Grutter et al. 2010; Sun et

al. 2012) and the presence of gnathiid isopods (the main food source of L. dimidiatus) is relatively

low in juvenile fishes (only 3.5 % of recruit P. amboinensis are infected) (Grutter et al. 2011). From

the cleaner wrasse’s perspective, these factors likely make settlers an unsuitable source of

ectoparasites. It was therefore unsurprising that we found that the damselfish that were cleaned by

juvenile L. dimidiatus were larger than the median size of non-cleaned conspecifics within a 1 m

radius. This result was consistent with previous studies that show that L. dimidiatus selectively

clean larger fish (Grutter et al. 2005; Clague et al. 2011), probably due to client size and parasite

load being closely correlated (Grutter 1994,1995a; Grutter and Poulin 1998); L. dimidiatus may

thus use size as an indicator of food availability.

From the client’s perspective, the odds of an individual new recruit (< 20 mm, TL) being

cleaned per 20 min, was low, with only two or three out of 1000 fish predicted to be cleaned when

there were no other fishes or recruits (< 20 mm) within a 1 m radius, and this rate rapidly decreased

81

towards zero with increasing abundance of nearby other fish species and recruits. Over a day,

however, the cumulative odds would be much higher. Such rare events could have

disproportionately significant benefits as some parasites can have especially harmful effects on

small individuals. While gnathiid isopods are generally considered micropredators, taking small

amounts of blood from their significantly larger host (Lafferty and Kuris 2002), they are capable of

consuming up to 85 % of the total blood volume of a settlement-stage P. amboinensis (Grutter et al.

2011). Indeed, infection by a single gnathiid isopod decreases the oxygen consumption and

swimming performance of a recruit damselfish and reduces the likelihood of successful settlement

on the reef (Grutter et al. 2011). If an infected fish has a gnathiid removed before it feeds on its

blood, this could provide a major benefit for survival. Since observed cleaning interactions were

brief (~ 1 s), it is unlikely that they obtained other direct benefit of cleaning, such as tactile

stimulation which leads to a reduction in client stress (cortisol) levels (Soares et al. 2011).

As direct cleaning interactions between cleaners and recently-settled recruits were relatively

uncommon, it is possible that the benefits of preferentially selecting a microhabitat near cleaners

may be at least partly indirect. For example, cleaner wrasse may reduce infection rate of parasites in

the vicinity of a cleaning station. As each L. dimidiatus eats around 1200 gnathiids per day (Grutter

1996a), this could significantly reduce the number of ectoparasites in the immediate vicinity of a L.

dimidiatus and hence lower the infection rate onto nearby settlers (Gorlick et al. 1987; Grutter

1999a; Clague et al. 2011). Other indirect benefits of the presence of cleaner wrasses may also exist

for settling fish. ‘Safe havens’ from predators may be created by L. dimidiatus (Cheney et al. 2008)

in that, in the presence of cleaners, some predators behave with notably reduced aggression. Settling

reef fish may therefore prefer to settle in a habitat with L. dimidiatus present, due to reduced risk of

predator aggression. Alternatively, or in addition, selection by young damselfish of microhabitats

near cleaners may simply be an investment where the return occurs only when recruits have grown

significantly.

On reefs lacking cleaner wrasse, the size at age of the lemon damselfish, P. moluccensis, is

smaller in older individuals (> 1 year old) and older fish have more parasitic copepods than fish on

reefs where cleaners are present (Clague et al. 2011). Additionally, size frequency distributions of

P. amboinensis and P. moluccensis are skewed towards smaller individuals on reefs without cleaner

wrasse (Waldie et al. 2011). Ultimately, this might affect fecundity, and hence the supply of future

settlers, since larger body sizes are correlated with egg production (Green 2008).

In summary, our study provides evidence that settlement-stage reef fish may use the

presence of cleaner wrasse as a cue when deciding what microhabitat to settle on. As the presence

of this mutualistic interaction creates favourable environments, this may ultimately enhance the

82

survival and abundance of these species. This study demonstrates that mutualism contributes to the

already-complex settlement processes of reef fishes.

Acknowledgements

We thank F. Cortesi, M. De Brauwer, F. R. A. Jaine, J. E. Lange, A. Marshell, G. A. C. Phillips,

and M. D. Mitchell for their help in the field; J. O. Hanson for statistical advice and R. Q-Y. Yong

and S. C. Cutmore for comments on earlier drafts of the manuscript. We are particularly grateful for

the support of the staff of the Lizard Island Research Station. This work was funded by a Sea World

Research and Rescue Foundation, Australia (SWRRFI) awarded to D. Sun, A. S. Grutter and T. H.

Cribb and funds from the Australian Research Council awarded to A. S. Grutter and M. G. Meekan

and from The University of Queensland awarded to A. S. Grutter.

83

Chapter 5: Diel patterns of infection by gnathiid isopods

on the ambon damsel, Pomacentrus amboinensis, and

the influence of the presence of cleaner wrasse

84

Chapter 5: Diel patterns of infection by gnathiid isopods on the ambon damsel,

Pomacentrus amboinensis, and the influence of the presence of cleaner wrasse

Derek Sun, Karen L. Cheney, Johanna Werminghausen, Thomas H. Cribb & Alexandra S. Grutter

Abstract

That the presence of bluestreak cleaner wrasse, Labroides dimidiatus, affects fish abundance and

fish traits is well known, yet little is known of the causal mechanism (s) involved. Since juvenile

gnathiid isopods are the main ectoparasite eaten by cleaner wrasse, a difference in their infection

number as a consequence of cleaner wrasse presence could be one such causal mechanism. To test

this hypothesis we used ambon damselfish, Pomacentrus amboinensis as a model host to estimate

the cumulative number of gnathiid isopods on patch reefs where the presence and absence of

cleaner wrasse had been experimentally manipulated for 12 years. Gnathiids were collected using

sentinel traps containing three live juvenile P. amboinensis as hosts; traps were sampled after 12 h

at dawn and dusk. There was no significant difference in the estimated cumulative number of

gnathiids per trap between reefs with and without L. dimidiatus. The probability of capturing one or

more gnathiid per trap on reefs without L. dimidiatus was 0.167 (95% CI = 0.112–0.240) compared

with 0.148 (95% CI = 0.096–0.221) on control reefs. However, the probability of capturing

gnathiids in traps was higher during the night [0.222 (95% CI = 0.144–0.327)], than during the day

[0.088 (95% CI = 0.061 – 0.126)]; or a 2.934 (95% CI = 1.725–4.985) times higher probability

during the night than during the day. We conclude that the rate of gnathiid isopod

attraction/detection measured here does not explain the patterns observed involving the effect of the

presence of L. dimidiatus on this fish reported in other studies.

Introduction

Parasites are often difficult to observe, yet many studies have shown that they may play a vital role

in influencing ecosystem functions (Hudson et al. 2006), regulate host populations (Scott and

Dobson 1989), and influence overall biodiversity (Hudson et al. 2006). On coral reefs, fish parasites

can have negative effects on their hosts (Ferguson et al. 2011; Binning et al. 2013; Krkošek et al.

85

2013). Understanding how fish manage parasite infections is thus critical to understanding their

ecology.

Cleaning behaviour, which involves organisms that remove ectoparasites from cooperating

clients, may be one way that reduces parasitic loads. Indeed, the frequent nature of cleaning

undergone by individual client fish, as often as every ~ 5 min for some species (Grutter 1995a),

suggests that the removal of ectoparasites is important to them. On Indo-Pacific coral reefs, L.

dimidiatus is the most common of the five Labroides species (Randall et al. 1997), all of which are

obligate cleaner fish that remove ectoparasites and sometimes tissue from the surfaces of other

fishes, known as clients.

The presence of the cleaner wrasse L. dimidiatus greatly influences the richness and

abundance of adult (Bshary 2003; Grutter et al. 2003), juvenile (Waldie et al. 2011) and recruiting

(Sun et al. in review, Chapter 3) reef fish. After 4 ̶ 20 months, in the Red Sea, the diversity of reef

fish on patch reefs was reduced when cleaner wrasse disappeared or were removed (Bshary 2003).

At Lizard Island, Great Barrier Reef, the species richness and abundance of visitors (fish species

that move between patch reefs) were reduced 18 months after cleaner wrasse had been removed

(Grutter et al. 2003); after 8.5 years, there was a reduction in the abundance and diversity of both

resident and visitor as well as adult herbivorous (Acanthuridae) fishes (Waldie et al. 2011). Only

recently, it was found that the abundance of damselfish recruits (species that are site-attached),

specifically Chrysiptera rollandi and Pomacentrus amboinensis, were reduced on these same patch

reefs after 12 years of regular removal of cleaner wrasse (Sun et al. in review, Chapter 3). These

studies highlight the importance of understanding the mechanism (s) via which a relatively

uncommon and small fish maintains the diversity and abundance of so many coral reef fishes.

An obvious way by which fish may benefit from the presence of cleaner fish is by obtaining

health benefits from a reduction in ectoparasites, which ultimately leads to changes in individuals or

communities. How ectoparasites can be influenced by cleaner fish will be determined by the

ecology and behaviour of the parasite and particularly the level of association between the parasite

and the host. There are two main types of ectoparasites, those that remain on the host for almost

their entire lives, i.e. ‘permanent’ parasites such as copepods (Finley and Forrester 2003) and

cymothoid isopods (Adlard and Lester 1994; Roche et al. 2013a), and those that remain on the host

only briefly while feeding, i.e. ‘temporary’ parasites such as gnathiid (Smit et al. 2003; Ota et al.

2012) and corallanid isopods (Grutter and Lester 2002), which are often classified as mobile

micropredators (Lafferty and Kuris 2002).

Surprisingly, only a few studies have tested the effect of L. dimidiatus on parasite loads.

The removal of L. dimidiatus did not have an effect on number of copepods per fish after 6 months

on a damselfish P. moluccensis at Lizard Island (Grutter 1996b). After 2 y the number of copepods

86

on P. vaiuli in the Marshall Islands was also not affected but the number of larger copepods (and

hence overall biomass) was higher (Gorlick et al. 1987). At Lizard Island, Clague et al. (2011)

showed that copepod number, summed across 10 fish sampled per reef, was higher on P.

moluccensis on reefs without a cleaner wrasse for 8.5 y, but only on larger P. moluccensis. While

copepods are not often consumed by cleaner wrasse (Grutter 1997) and individual damselfish are

cleaned relatively infrequently (Grutter 1995a), that larger P. moluccensis individuals are

preferentially cleaned (Clague et al. 2011; Sun et al. in review, Chapter 4) could explain why an

effect on copepod number was only apparent in larger individuals.

The presence of cleaners also has an effect on parasitic isopods. For caged thicklip wrasse,

Hemigymnus melapterus, gnathiid number per fish was higher on reefs without L. dimidiatus after

24 h and 12 d (Grutter 1999a), while on the same removal reefs the prevalence and number of

corallanid isopods was higher after 24 h, while a noticeable shift towards smaller isopods was

detected after 12 d (Grutter and Lester 2002). In a very different system involving cleaner gobies

(Elacatinus spp.) in the Caribbean, gnathiid number for Stegastes diencaeus damselfish captured

using handnets was lower on individuals that had cleaning stations within their territory than on

individuals that lived further away from a cleaning station (Cheney and Côté 2001).

Gnathiids can inhibit the growth (Jones and Grutter 2008), swimming performance and

oxygen consumption (Grutter et al. 2011) and decrease the survival rate (Grutter et al. 2008; Grutter

et al. 2011) of damselfish juveniles. High numbers of gnathiids can cause anaemia (Paperna 1977;

Grutter 2008) and mortality in adult fish (Mugridge and Stallybrass 1983; Marino et al. 2004).

Thus, a reduction in gnathiids could ultimately lead to the higher abundance of damselfishes

observed on reefs where cleaner wrasse are present (Bshary 2003; Waldie et al. 2011; Sun et al. in

review, Chapter 3).

The size frequency distribution of both P. amboinensis and P. moluccensis was shifted

towards smaller individuals on reefs where cleaner wrasse have been removed for 8.5 years (Waldie

et al. 2011). The shift in size distribution could be due to decreased rate of growth in individuals on

reefs without a cleaner wrasse. Although P. moluccensis had a lower growth rate and higher

copepod numbers on reefs without cleaners, this was only apparent in larger individuals (Clague et

al. 2011). Thus, it is possible that other ectoparasites such as gnathiid isopods may explain the

overall shift in size distribution seen between reefs with and without cleaner wrasse.

As L. dimidiatus can consume an average of 1200 ectoparasites a day with the majority of

their diet being made up of gnathiid isopods (99.7 %) (Grutter 1996a), it is possible that their effect

on parasites is likely to be the greatest on this group. Gnathiid isopods are one of the main

ectoparasites that infect coral reef fishes (Grutter et al. 2000). Unlike copepods, gnathiids are

mobile temporary parasites, emerging either during the day or at night to locate a host to feed

87

(Sikkel et al. 2006). Gnathiids usually detach from a host once they are fully fed (Grutter 2003) or

are disturbed (Grutter 1995b). Due to this unique life cycle and behaviour, several measures have

been used to quantify the ‘abundance’ of gnathiids. These terms are defined as density of

cumulative number of emerging gnathiids (i.e. number per reef area/per time), number (number per

fish) and infestation (number present per unit time). Therefore, the sampling method used to

quantify gnathiid isopod ‘abundance’ on fish could also affect the estimate; for example due to

differences in their emergence from the benthos (Grutter 1999b; Sikkel et al. 2006), the time of day

that a fish are sampled will affect how many parasites are counted. Damselfishes also tend to have

few gnathiids (Grutter 1994; Sun et al. 2012) and so the odds of sampling them on fish are low. A

method that increases the time they are sampled on fish would increase these odds.

The use of caged fish in previous studies to quantify the number or infestation of gnathiids

(Grutter 1999a; Sikkel et al. 2006; Coile and Sikkel 2013) does not exclude the potential direct

interaction between cleaner and the caged fish and hence the removal of gnathiids. The use of cages

also does not prevent the potential loss of gnathiids once they are fed and have dropped off.

Therefore, it is possible that the initial number of gnathiids in previous studies was similar on caged

fish placed on reefs either with or without cleaner wrasse and that a lower infection was recorded

simply due to the direct cleaning interaction between cleaners and the caged fish. As gnathiids can

easily detach themselves from a host, conventional methods of capturing fish (handnets) to examine

the number of gnathiids are not ideal (Grutter 1995b). Such detachment could have contributed to

the low prevalence rate and number of gnathiids that has been reported on damselfishes (Grutter

and Poulin 1998; Cheney and Côté 2001; Grutter et al. 2011; Sun et al. 2012).

The high emergence rates of gnathiids (41.7 ± 6.9 m-2 day-1) (Grutter 2008) means it is

likely that hosts are put under constant physical stress, as well-fed gnathiids are constantly replaced

by unfed individuals. Although previous studies have shown an effect of cleaner fish presence on

the number of gnathiids on a fish (Grutter 1999a; Cheney and Côté 2001), the results are limited in

that they only provide information of the number of gnathiids on a host at a given time, when the

host was captured. As a result, these studies may not provide a true representation of overall number

in relation to cleaner fish presence. Therefore, an alternative sampling technique is required: one

which prevents cleaner wrasse from interacting with the host fish, retains gnathiids once they have

dropped off, and provides as estimate of the cumulative number of gnathiids on fish over time. One

such method involves the use of ‘sentinel traps’ which hold ‘bait’ fish and allow gnathiids, but not

cleaner fish, to enter and feed off hosts, but which do not allow gnathiids to escape (Sikkel et al.

2011).

Despite damselfishes being relatively sedentary on coral reefs and unable to move between

reefs to seek a cleaner wrasse, their abundance is higher in the presence of a cleaner wrasse (Sun et

88

al. in review, Chapter 3). Yet, the mechanism(s) remains unclear. Currently, it is not known if the

cumulative number of gnathiids on damselfishes differ according to cleaner wrasse presence. As

cleaner wrasse can consume large quantities of gnathiid isopods, we therefore predict that the

cumulative number of gnathiids will be lower on reefs where cleaner wrasse are present. In this

study, we examined whether the presence of cleaner wrasse has an effect on the cumulative number

of gnathiid isopods per trap using relatively large juvenile ambon damsel, P. amboinensis.

Pomacentrus amboinensis were placed in sentinel traps and placed on experimental patch reefs with

or without cleaner wrasse. Because cleaner fish are only active during the day (Grutter 1996a) and

thus could potentially affect only ‘diurnal gnathiids’, sampling was divided between the day and the

night, over a 24 h time period.


Study site

This study was conducted at Lizard Island (1440S, 14528E) in the northern GBR, Australia during

late summer (January and February) of 2013. We used 16 isolated patch reefs (3−7 m depth) located

at two sites: the Lagoon (11 reefs) and at Casuarina Beach (5 reefs) (Grutter et al. 2003; Clague et al.

2011). Reefs were randomly allocated as either removal or control treatment reefs in September 2000.

Subsequently, reefs were checked at several-month intervals and were inspected for cleaner wrasse

L. dimidiatus; any new recruits on removal reefs were removed using barrier and handnets. Control

reefs were surveyed periodically for L. dimidiatus by swimming around the reef several times and the

abundances of cleaners recorded.

Study species

The ambon damselfish, Pomacentrus amboinensis, is a common coral reef fish species on the GBR.

As much is known about its ecology, including habitat preference, recruitment, and feeding habits

(Kerrigan 1996; McCormick 2006; McCormick et al. 2010), this species has been extensively used

as a model species in both field and laboratory experiments and it is abundant on all of the

experimental reefs.

Sampling design

A total of 72 large juvenile P. amboinensis (mean TL per fish ± SE = 50.01 ± 0.38 mm) were

collected at Lizard Island in January 2013 using barrier and handnets. Fish were initially maintained

in aerated 40 L tanks, then, due to their known tendency for aggressive behaviour towards each

other, were transferred to individual clear plastic holding aquaria (29 × 17 × 12 cm) with a constant

89

flow of unfiltered seawater and maintained on a diet of commercial fish flakes. Fish were then

randomly divided into four groups (A, B, C, and D), each group consisting of 18 fish. Within each

group, the fish were then further divided into 6 labelled subgroups (numbered from 1–24) with 3

fish in each subgroup. These three fish were used as the sampling unit per fish trap. Gnathiids

mainly locate their hosts using olfactory cues (Sikkel et al. 2011) so three fish per trap were used in

order to maximise the amount of chemical cue emitted from the trap.

A total of 14 individually labelled sentinel traps were constructed using a T-junction PVC

pipe (370 mm long × 150 mm diameter) with a threaded lid attached to each of the three ends

(Figure 1) (see Sikkel et al. 2011 for a base design). A 100 mm diameter hole was cut out of each

lid and a clear plastic funnel (10 mm diameter opening) attached to the inside surface of the lid,

using plastic cable ties and aquarium grade silicone glue. To facilitate filling and emptying the

water in the tubes, three 32 mm diameter holes were drilled into the sides of the tube. The drainage

holes were covered with 100 μm plankton mesh glued inside the tube; this allowed only water (no

gnathiids) to circulate in and out of the trap. The six openings allowed sufficient water circulation to

maintain adequate oxygen levels.

Twelve traps in a paired design, with 6 traps placed each time on a removal and control

reef, were used per day. This was repeated on all 16 experimental reefs; due to the odd number of

reefs at each of the two sites sampled, this meant only one reef was sampled on one day at each of

the sites, resulting in 9-day sampling periods. For each sampling day, two fish groups (A and B, or

C and D) were used. The traps were placed on the reef at dusk, sampled (traps emptied and

redeployed with same fish) at dawn the following day (~ 12 h after initial placement) and the

second and final sample collected at dusk later that day (~ 12 h after first sampling). The sampling

design allowed each group to be used a total of 4 different times throughout the experiment, with

each group being re-used every 48 h. This provided each group with a recovery period of at least 24

h between deployments.

To determine whether gnathiids were attracted to traps and not the fish in them, we

deployed empty traps as a control. Empty trap sampling was conducted with 6 traps placed on a

control and 6 on a removal reef, one day before the commencement (day 0) and the day after the

conclusion (day 10) of the 9-day experiment. As an additional control, every second day throughout

the experiment, an extra empty trap was placed alongside the six traps containing fish. The entire

experiment was replicated after a one-week recovery period, with the same groups of fish, in the

same order, for the same duration (Table 1).

90

Table 1 Number of traps (with and with no fish) sampled per sampling day, site, reef pair cleaner

presence treatment, and day and night. Total n per host = 382 traps.

Experiment

round

Sampling

day

Site Reef pair cleaner

treatment

Number of traps with

fish

(Control/Removal)


no fish

(Control/Removal)

Day Day Day Night

One 0 Casuarina

Beach Control/Removal - - 6/6 6/6

One 1 Casuarina

Beach Control/Removal 6/6 6/6 0/1 0/1

One 2 Casuarina


One 3 Casuarina

Beach Control 6/6 6/6 0/1 0/1

One 4 Lagoon Control/Removal 6/6 6/6 1/1 1/1





One 9 Lagoon Control 6/6 6/6 0/1 0/1

One 10 Lagoon Control/Removal - - 6/6 6/6

Two 0 Casuarina

Beach Control/Removal - - 6/6 6/6

Two 1 Casuarina


Two 2 Casuarina


Two 3 Casuarina

Beach Control 6/6 6/6 0/1 0/1

Two 4 Lagoon Control/Removal 6/6 6/6 1/1 1/1


Two 6 Lagoon Control/removal 6/6 6/6 1/1 1/1



Two 9 Lagoon Control 6/6 6/6 0/1 0/1

Two 10 Lagoon Control/Removal - - 6/6 6/6

91

Figure 1 Photograph of the sentinel trap used in the experiment. Traps were made from plastic PVC

material. A) Each lid is fitted with a transparent plastic funnel, allowing gnathiid isopods to enter

the trap while providing fish with adequate water circulation; B) Each trap is fitted with a 1.4 kg

dive weight, which is secured with a dive belt at the front opening; C) Three drainage holes (3.2 cm

in diameter) are drilled at the back of the traps. Photo: Maarten De Brauwer.

Sampling protocol

The three fish were held individually in subcages within the trap. Subcages reduced any

potential negative interaction amongst the fish within the trap and were constructed from flexible

corrugated plastic pipe (100 diameter × 50 length mm) with mesh (15 mm) covering each circular

end (Figure 2). Although mesh size was large enough to allow gnathiids to enter the subcage, ~10

larger holes (3 mm) were also drilled in the mesh to increase the opportunity for gnathiids to enter

the subcage. The subcages had small holes bored along its length in order to increase the entry

points for gnathiid isopods. Prior to transporting fish to the field site, each fish was placed

individually into subcages labelled with the corresponding subgroup identification.

Fish were transported to the field in a 60 L container, which was aerated with regular water

changes. At the field site, each subgroup was placed in a trap. Traps were then passed to a snorkeler

and were placed carefully on the reef at dusk. Traps were secured with a 1.4 kg dive weight

attached to the outside of each trap using a dive belt and buckle. Traps were subsampled at dawn

the following day. To ensure no trapped gnathiids were lost while moving traps off the reef to the

boat, all funnel holes were blocked using a 1 cm diameter ball of household gum (Blu-tack,

Thomastown, Australia). The contents of each trap were emptied and well-rinsed into

correspondingly-numbered 10 L buckets. Subcages, still containing a fish, from each trap were

given a 1-minute freshwater bath (3 L) in a separate bucket, and the contents then correspondingly

B

C

A

92

placed into the aforementioned 10 L buckets. Freshwater baths are a standard method used to

remove gnathiids that may be attached on a fish (Soares et al. 2008). Subcages and fish were then

returned to the traps and placed on their original locations on the reef. This procedure was repeated

the next dusk but using the next pair of reefs and the next two fish groups.

Twelve buckets holding the trap, rinse and bath water samples were brought back to the

research station, along with (after the dusk sampling) the recently-used fish in their subcages. For

the dusk samples, an airstone was placed into each bucket for 3 hours to keep the fish alive while

allowing any remaining gnathiids to drop off naturally. Each fish subgroup was also given a final 1-

minute freshwater bath, during which their body was gently rubbed before being placed back into

their allocated aquaria. The combined water samples were filtered through a 100 μm plankton sieve

into vials to remove any gnathiids, and then preserved with a 10% formaldehyde solution in

seawater. During the experiment a total of six fish died (two traps, each on a control and removal

reef) and were replaced with new ones.

Figure 2 Individual subcages used to hold Pomacentrus amboinensis in the sentinel traps

Statistical analysis

Since 53 of the 62 samples with gnathiids present contained only one individual, we used the

presence/absence of gnathiids as the response variable. A generalized linear mixed effect model

(GLMER) with a binomial distribution error was used to examine whether gnathiid presence

differed with cleaner wrasse presence. The terms treatment (cleaner or no cleaner), site (Lagoon or

Casuarina Beach), time (day or night) and experiment replicate (Experiment one or two) were fixed

factors, and reef ID, sampling day, and fish group as random factors. Fish subgroup identity was

nested with fish group. The sum of the fish weight of the three fish per subgroup was included as a

covariate, as gnathiids are known to use fish scent to locate a potential host (Sikkel et al. 2011) and

we expected that the scent of fish would be correlated with the mass of fish.

93

In a separate analysis, the same model was used to estimate the rate at which gnathiids

entered empty traps (no fish), but without fish weight, fish group, and fish subgroup identity as

factors.

To determine the best model for the data, we compared the full model with models in

which one of the explanatory terms was dropped using the “drop1” function (Chambers 1992). If

the maximum likelihood ratio test (LRT) found that a dropped term had no significant effect on the

model by using the Chi-square distribution, then the term was dropped. All statistical analyses were

performed using R version 3.0.1 (R Development Core Team 2013). As information regarding

species-level taxonomy of gnathiid isopods is scarce, gnathiids encountered in this study were not

identified further.

Results

Diel patterns

Of the 382 trap samples containing fish, 62 contained a total of 72 gnathiids, with a maximum of

three gnathiids in a single trap (Table 2).

The presence of gnathiid isopods differed according to time of day (LRT= 13.61, df = 1, P

= 0.0002), with more gnathiids being caught at night than in the day. Traps at night had a higher

probability of capturing a gnathiid 0.222 (95% CI = 0.144–0.327), than those set during the day

0.088 (95% CI = 0.061–0.126); or a 2.934 (95% CI = 1.725–4.985) times higher probability during

the night than during the day.

Cleaner wrasse presence

The simplified model showed that the presence/absence of gnathiid isopods did not differ according

to cleaner wrasse presence (LRT = 0.216, df = 1, P = 0.642), site (LRT =0.050, df = 1, P = 0.822),

fish weight (LRT = 1.185, df = 1, P = 0.276) or experiment replicate (LRT = 0.299, df = 1, P =

0.583). Traps had a 0.167 (95% CI = 0.112–0.240) probability of capturing a gnathiid on removal

reefs, compared with 0.148 probability (95% CI = 0.096–0.221) on control reefs, or a 1.153 (95%

CI = 0.731–1.818) times higher probability on removal than control reefs.

Of the 128 trap samples containing no fish, 11 contained a sum of 12 gnathiids, with a

maximum of two gnathiids per trap (Table 2). The presence/absence of gnathiids did not differ

according to cleaner wrasse presence (LRT = 0.136, df = 1, P = 0.711), time of day (LRT = 0.239,

df = 1, P = 0.624), site (LRT = 2.43, df = 1, P = 0.118) and experiment replicate (LRT = 1.22, df =

1, P = 0.268).

94

Table 2 Number of sentinel traps with gnathiids, day and night trap samples and the number of gnathiids per trap collected from the sentinel trap

experiment on Lizard Island.


isopods (day/night)

Number of fed/not

fed gnathiids # Gnathiid isopods per trap

Treatment

Total

number of

traps

Traps with

gnathiids

Day Night Fed Not fed 0 1 2 3

Fish traps

Control

382

33

11

22

30

12

181

25

7

1

Removal 29 7 22 22 8 139 28 1 0

No fish traps

Control 128

6 3 3 2 4 58 6 0 0

Removal 5 3 2 2 4 59 4 1 0

95

Discussion

Our study found that the cumulative number of gnathiid isopods per trap using the three ambon

damsel did not differ between patch reefs that have cleaner wrasse present and those from which

cleaners had been experimentally removed over a 12-year period. There was a clear difference in

the number of gnathiids according to day and night. The results from this study suggest that factors

related to cleaner wrasse presence other than gnathiids alone in this system may explain the benefits

that occur to damselfishes.

Diel patterns

Although the number of gnathiids did not differ with cleaner wrasse presence or absence, it

did differ according to day or night. There was a 2.9 times higher probability that traps sampled at

night would capture a gnathiid than traps that where sampled during the day. The diel variation seen

in this study is consistent with findings using other sampling methods, which have also reported

that the number of gnathiids on caged fish is higher at dawn than at dusk (Grutter and Hendrikz

1999; Sikkel et al. 2006). Although, gnathiids clearly can emerge and infect fish during the day, this

is usually at a much lower rate than during the night, with peak activity being recorded at night and

at pre-dawn periods (Grutter and Hendrikz 1999; Sikkel et al. 2006).

The higher abundance of gnathiid isopods active at night may have arisen as a result of

selective pressures from diurnal parasite predators, principally cleaner fish, that feed on them

during the day (Grutter 2002). As different ontogenetic stages (Grutter et al. 2000; Sikkel et al.

2006) and species (Jones and Grutter 2007) of gnathiids are known to emerge differentially during

either the day or at night, it is possible that some species have evolved to emerge at night when

there are fewer diurnal planktivores, cleaners and microinvertebrate feeders (Helfman 1986; Grutter

1996a), therefore, reducing the risk of predation. Although predation of gnathiid isopods can still

occur from nocturnal cleaning activities as observed in the cleaner shrimp, Urocaridella sp. on coral

reefs (Bonaldo et al. 2014) and on monogeneans by Lysmata amboinensis in tanks (Militz and

Hutson 2015), gut contents analyses reveal that cleaner shrimps consume few gnathiids (Becker and

Grutter 2004). Therefore, it is unlikely that cleaning activities by shrimps will have a major impact

on the abundance of gnathiids, compared with the cleaning activities of cleaner fish.

Cleaner wrasse presence

We found that there was no difference in the probability that a gnathiid was present in a trap

on reefs with (0.148) and without cleaner wrasse (0.167). As gnathiid presence did not differ

according to cleaner wrasse presence, it may therefore not explain some of the positive effects

observed on fish from reefs with cleaner wrasse present. For example, the smaller size distribution

96

(Waldie et al. 2011) of P. amboinensis and P. moluccensis and reduced growth rate (Clague et al.

2011) reported in earlier studies which used the same experimental reefs may not be due to

differing levels of gnathiids on these fishes. Indeed Clague et al. (2011) showed that the number of

copepods was higher on larger P. moluccensis from reefs that have had cleaners removed than on

reefs with cleaners and suggested this could explain their reduced growth rate. There is evidence

that infection by copepods can affect the growth, gonad mass and survival of fish (Finley and

Forrester 2003; Palacios-Fuentes et al. 2012). Other ectoparasites such as cymothoid isopods have

also been shown to affect growth and survival in reef fishes (Adlard and Lester 1994; Fogelman and

Grutter 2008). It is thus possible that other ectoparasites such as copepods and cymothoid isopods,

in isolation or even in synergy with gnathiids, may cause this downward shift in size. However, we

did not find any cymothoids in the traps. Similarly, the recent study that showed that the abundance

of recruiting damselfishes was higher on reefs where cleaner wrasse are present (Sun et al. in

review, Chapter 3) is thus unlikely due to higher gnathiid numbers.

Both the prevalence and number of gnathiid isopods was low, with only 16 % of traps over a

24 h time period having at least one gnathiid, with the number of gnathiids ranging only from one to

three gnathiids per trap. This low prevalence and number of gnathiids is consistent with previous

studies that demonstrated that small juvenile P. amboinensis (mean TL ± SE = 14.28 ± 0.26) are

rarely infected with gnathiids, with instantaneous prevalence being 3 and 3.5 % and never more

than one gnathiid per infected fish (Grutter et al. 2011; Sun et al. 2012). The feeding success of a

micropredator should increase as host mass increases, as the likelihood of being detected and thus

potentially being preyed upon decreases (Kuris and Lafferty 2000). This could explain the lower

number of gnathiids we recorded infecting the relatively small-sized P. amboinensis.

Due to the low prevalence of gnathiids infecting these fish, it is unlikely that gnathiids are

the likely mechanisms that have caused the higher abundance of damselfish recruits on reefs with

cleaner wrasse than on reefs without. Therefore, it is likely that other factors related to cleaner

wrasse presence could explain those results. For example, decreased aggression from predators

when in the presence of cleaner wrasse (Cheney et al. 2008) or reduced stress via tactile stimulation

provided by cleaner wrasse (Soares et al. 2011).

It is also possible that we could not detect a difference due to cleaner wrasse presence with

the sentinel traps due to the sampling method. The amount of sampling time (24 h) may have been

too limited to detect slight differences in gnathiid abundance. Convincingly, further sampling would

detect a small yet significant difference in gnathiid abundance between cleaned and non-cleaned

reefs, such a small difference might still be important. We emphasize that although small fish may

only be attacked by a few gnathiids over its life time this might be enough to reduce their growth

and potentially affect their survival.

97

A speculative scenario that agrees with our results as well as the observed beneficial effects

on fishes, is that the advantage to fish for being on reefs with L. dimidiatus could be due to the

length of infection (feeding period) being reduced, rather than the number of gnathiids attacking a

fish. This could occur if cleaners disturb gnathiids and reduced the time they spend (and the blood

they remove) on fish. Previous studies have shown that the feeding rate (time spent ingesting blood)

of gnathiid isopods can vary among individuals, with gnathiids remaining on a host for as little as

30 min or up to 10 h in order to feed (Grutter 2003; Smit et al. 2003; Grutter et al. 2011).

The efficiency of L. dimidiatus in seeking and consuming gnathiids, coupled with the

frequency at which clients are cleaned (e.g. barred rabbit fish, Siganus doliatus, are cleaned on

average every 5 min in a day), means that the risk of predation can be exceedingly high for some

gnathiids (Grutter 1995a). Grutter (2003) showed that for third-stage larval gnathiids, the likelihood

of red blood being present in the gut increased over time, with 20 % of gnathiids being engorged at

15 minutes and 46 % engorged at 30 min. Given that clients such as S. doliatus are cleaned

frequently, the amount of red blood that they may lose from gnathiids should be significantly

reduced on reefs with cleaner wrasse. Gnathiids may not be able to become fully engorged on red

blood within the shortened time frame, due to the client being cleaned (gnathiid predation). This

scenario fits the paradigm of cleaner wrasse driving changes in gnathiid feeding behaviour,

ultimately limiting their ability to adversely affect their host and thus conferring benefits upon fish

clients.

Although studies have shown a clear size preference by L. dimidiatus for larger host

individuals (Grutter 1995a; Grutter et al. 2005; Clague et al. 2011), it should be emphasised that a

reduction in parasitic numbers would be beneficial to any host, regardless of how frequently it may

be cleaned. A single gnathiid can consume up to 85 % of the total blood volume of a settlement-

stage damselfish (Grutter et al. 2011) and might transmit blood parasites and other infectious agents

(Davies et al. 2009; Curtis et al. 2013). Therefore, fish species that are cleaned less regularly (e.g. P.

moluccensis, which are only cleaned on average once per 30 min) (Grutter 1995a) would still

benefit from a reduction in gnathiid feeding time due to a reduced blood loss and exposure to

parasite transmission.

Previous studies have clearly demonstrated that there is an advantage for fish being on reefs

with cleaners present (e.g. Grutter 1999a; Soares et al. 2011; Waldie et al. 2011). In this light the

results of this study showing no differences in relation to cleaner wrasse presence are surprising. In

consideration, these results suggest that the benefits gained from being on reefs with cleaner wrasse

(such as increased recruitment), could relate to the minimisation of the infection that do occur. In

summary, this study demonstrates that the relationship between cleaner wrasse and their clients is

98

complex and that factors other than the abundance of gnathiid isopods in the environment may

explain some of the advantages to fish of being on reefs where cleaner wrasse are present.

Acknowledgements

We thank E. C. McClure, M. De Brauwer, F. Cortesi, G. A. C. Phillips and A. E. Winters for their

help in the field; S. P. Blomberg for statistical advice, P. C. Sikkel for advice on sentinel trap and

experimental design and R. Q-Y. Yong for comments on earlier drafts of the manuscript. We are

particularly grateful for the support of the staff of the Lizard Island Research Station. This work was

funded by a Sea World Research and Rescue Foundation, Australia (SWRRFI) awarded to D. Sun,

A. S. Grutter and T. H. Cribb and funds from the Australian Research Council awarded to A. S.

Grutter.

99

Chapter 6: General Discussion

100

Chapter 6: General Discussion

The Preamble to this thesis referred to the complexity of interactions between three groups of

organisms – hosts, parasites and cleaners. Although a lot of work has been done on the various

elements of this system prior to my study, and although substantial progress has been made here,

the questions arising from the study appear to leave at least as much to be done as when I started.

From the study of the diversity of digenean trematodes of adult pomacentrid fishes at Lizard

Island it was established that digenean richness is low relative to that of other coral reef fish groups.

Host-specificity ranges dramatically from a few taxa that are strongly oioxenic through to several

that are clearly stenoxenic and others that are seemingly euryxenic. In comparison with the known

fauna of Heron Island, the fauna at Lizard Island was generally comparable but with several key

distinctions. Questions arising from these findings start with the problems of identification. It was a

surprise to find evidence of so much cryptic richness among specimens of just one of the 19

morphospecies in the study. Hysterolecitha nahaensis clearly is deserving of and requires dedicated

independent study. There is no a posteriori explanation for why this one species has radiated so

dramatically. In terms of the discovery of the total richness of the system, the distribution of

oioxenic species presents a challenge. Not only do we not understand why only one faustulid is

present in one species of Chromis, one bivesiculid in Acanthochromis polyacanthus and Lepotrema

adlardi in just one species of Abudefduf – we are left uncertain as to what richness may remain

undetected in the pomacentrid species that have never been examined. In terms of beta diversity, it

might have been predicted that, as found, there would be some relatively low-level differences

between the fauna found at Heron and Lizard Islands. What remains completely unexplained is why

the species that have limited distributions are so restricted thus. Further, the striking differences in

mean richness and mean prevalence between Heron and Lizard Islands remain unexplained. For

convincing explanations to emerge on such topics we will need at least to understand the life-cycles

of these taxa. Presently not a single life-cycle of a pomacentrid-infecting digenean is known.

This study has also highlighted the importance of the presence of cleaner wrasse in the

ecology of young reef fishes. I demonstrated that the positive effects of cleaner wrasse presence on

adult fish population size reported in other studies (Bshary 2003; Grutter et al. 2003; Waldie et al.

2011), begins during the recruitment period by enhancing the abundance of damselfish recruits.

Thus, we now know that cleaner wrasse can influence the population dynamics from the start of the

recruitment period. Although cleaner wrasse species are important key stone mutualist on coral

reefs, they are collected in large numbers for the aquarium trade; 20,000 individuals have been

caught in Sri Lanka in one year alone. The removal of such a key organism in large numbers should

101

have a detrimental effect on the biodiversity of reef fishes. Therefore, studies on reef fish recovery

on coral reefs should consider the beneficial role of cleaner wrasse alongside that of other key

organisms. However, it remains unclear why the abundance of only two species of damselfishes

(Chrysiptera rollandi and Pomacentrus amboinensis) which differ from each other in regards to

ecology (e.g. habitat and social structure), were affected by the presence of a cleaner wrasse, when

similar species showed no differences. It is possible that factors other than the presence of cleaner

wrasse dominate when it comes to microhabitat selection during their recruitment. Despite the

overall higher abundance of recruiting damselfishes on reefs where cleaner wrasse are present, it is

not understood how these fish are selecting these particular reefs. As cleaner wrasse are inactive and

shelter within the reef matrix at night when the majority of settlement-stage fish settle, it is possible

that they are selecting reefs based on olfactory in addition to visual cues. This hypothesis is yet to

be tested, however, doing so will provide greater insight into the dynamics that occur during the

recruitment period.

The results from chapter 4 showed that when given the option, a young damselfish would

preferentially select a microhabitat that is adjacent to a cleaner wrasse. This finding indicates that

cleaner wrasse may influence small-scale post-settlement migration. Field observations indicated

that small recruits are rarely cleaned by juvenile cleaner wrasse. It was therefore unsurprising that

the damselfish that were preferentially cleaned by juvenile L. dimidiatus were larger. It is thus

likely that indirect rather than direct benefits are the driver for such small recruits seeking proximity

to a cleaner. As cleaner wrasse are known to consume large quantities of parasites per day such as

gnathiids, a reduction in gnathiid abundance within cleaning stations of juvenile cleaner wrasses

could be one possible benefit. However, it is not yet known if the abundance of gnathiids are

actually reduced within cleaning stations. Thus, further studies are required to explore this scenario.

It is also not clear what other indirect benefits a recruit may obtain from being in a microhabitat that

is near a cleaner wrasse. If gnathiid infestation is lower in the vicinity of a cleaner wrasse, we

should expect both growth and survival to increase, thus possibly explaining the results found in

this study. However, alternatively, or in addition, selection by young damselfish of microhabitats

near cleaners may simply be an investment where the return only occurs when recruits have grown

significantly.

Given that past studies have shown reduction in gnathiid infestation on hosts and that

cleaner wrasse are known to consume enormous numbers of gnathiids when cleaner wrasse are

present, it was somewhat surprising to find no effect of cleaner wrasse presence in relation to the

abundance of gnathiids caught in sentinel traps. As this study only examined the infection of one

species of damselfishes, future studies should examine whether the infection of gnathiids is

consistently the same or different for other species of damselfishes. It is possible that different

102

species of damselfishes may be infected with more gnathiids. However, the results shown in this

study may suggest that the benefits of fish being on reefs with cleaner wrasse are not significantly

from a reduction in gnathiid isopod load but due to exposure of gnathiid infection being minimised

by cleaner wrasse presence. Further examination is required to test this idea. If supported, the

implications could be an intriguing “minimal effect/win/win” outcome whereby the population of

gnathiids is not materially affected by the presence of cleaners, the cleaners benefit from their

presence by the provision of most of their food, and client fish benefit from the presence of the

cleaners by the reduction of the impact of gnathiid parasitism. Could there be further complexities

such as cleaners preferentially eating engorged gnathiids? Only further studies will tell.

103

References

Adam TC (2010) High-quality habitat and facilitation ameliorate competitive effects of prior residents on new settlers.

Oecologia 166:121-130

Adamson ML, Caira JN (1994) Evolutionary factors influencing the nature of parasite specificity. Parasitology

109:S85-S95

Adlard RD, Lester RJG (1994) Dynamics of the interaction between the parasitic isopod, Anilocra pomacentri, and the

coral reef fish, Chromis nitida. Parasitology 109:311-324

Aiken HM, Bott NJ, Mladineo I, Montero FE, Nowak BF, Hayward CJ (2007) Molecular evidence for cosmopolitan

distribution of platyhelminth parasites of tunas (Thunnus spp.). Fish Fish 8:167-180

Allen GR (2008) Conservation hotspots of biodiversity and endemism for Indo-Pacific coral reef fishes. Aquat Conserv

18:541-556

Almany GR (2003) Priority effects in coral reef fish communities. Ecology 84:1920-1935

Almany GR, Webster MS (2006) The predation gauntlet: early post-settlement mortality in reef fishes. Coral Reefs

25:19-22

Anderson GR, Barker SC (1998) Inference of phylogeny and taxonomy within the Didymozoidae (Digenea) from the

second internal transcribed spacer (ITS2) of ribosomal DNA. Syst Parasitol 41:87-94

Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA+ for Primer: Guide to software and statistical methods,

Plymouth

Arnal C, Côté IM (2000) Diet of broadstrip cleaning gobies on a Barbadian reef. J Fish Biol 57:1075-1082

Arnal C, Morand S (2001) Importance of ectoparasites and mucus in cleaning interactions in the Mediterranean cleaner

wrasse Symphodus melanocercus. Mar Biol 138:777-784

Barber I, Hoare D, Krause J (2000) Effects of parasites on fish behaviour: a review and evolutionary perspective. Rev

Fish Biol Fish 10:131-165

Barker SC, Bray RA, Cribb TH (1993) Preptetos cannoni n. sp. (Digenea: Lepocreadiidae) from Siganus lineauts

(Teleostei: Siganidae) from the southern Great Barrier Reef, Australia. Syst Parasitol 26:151-155

Barker SC, Cribb TH, Bray RA, Adlard RD (1994) Host-parasite associations on a coral reef: Pomacentrid fishes and

digenean trematodes. Int J Parasitol 24:643-647

Barnett LJ, Miller TL, Cribb TH (2014) A review of the currently recognised zoogonid cercariae, including the

identification and emergence ecology of Cercaria capricornia XI (Digenea: Zoogonidae) from Nassarius

olivaceus (Gastropoda: Nassariidae) in Central Queensland, Australia. Folia Parasitol (Praha) 61:322-330

Bartoli P, Gibson DI (2007) Synopsis of the life cycles of Digenea (Platyhelminthes) from lagoons of the northern coast

of the western Mediterranean. J Nat Hist 41:1553-1570

Bartoli P, Gibson DI, Bray RA (2005) Digenean species diversity in teleost fish from a nature reserve off Corsica,

France (Western Medterranean), and a comparison with other Mediterranean regions. J Nat Hist 39:47-70

Bartoli P, Bray RA, Gibson DI (2006) Four closely related but forgotten species of Rhipidocotyle Diesing, 1858

(Digenea : Bucephalidae) in fishes from European seas. Syst Parasitol 65:129-149

Bascompte J, Jordano P (2007) Plant-animal mutualistic networks: The architecture of biodiversity. Annu Rev Ecol

Evol Syst 38:567-593

Bay LK, Jones GP, McCormick MI (2001) Habitat selection and aggression as determinants of spatial segregation

among damselfish on a coral reef. Coral Reefs 20:289-298

Beasley VR, Faeh SA, Wikoff B, Eisold J, Nichols D, Cole R, Schotthoefer AM, Staehle C, Greenwell M, Brown LE

(2005) Risk factors and the decline of the northern cricket frog, Acris crepitans: evidence for involvement of

herbicides, parasitism, and habitat modifcations. In: Lanoo MJ (ed) Amphibian declines: the conservation

status of United States species. University of California Press, Berkeley, CA,

Becker JHA, Grutter AS (2004) Cleaner shrimp do clean. Coral Reefs 23:515-520

Becker JHA, Curtis LM, Grutter AS (2005) Cleaner shrimp use a rocking dance to advertise cleaning service to clients.

Curr Biol 15:760-764

Biani NB, Mueller UG, Wcisio WT (2009) Cleaner mites: Sanitary mutualism in the miniature ecosystem of neotropical

bee nests. The American Naturalist 173:841-847

Binning SA, Roche DG, Layton C (2013) Ectoparasites increase swimming costs in a coral reef fish. Biol Lett

9:20120927

Blair D, Bray RA, Barker SC (1998) Molecules and morphology in phylogenetic studies of the Hemiuroidea (Digenea:

Trematoda: Platyhelminthes). Mol Phylogenet Evol 9:15-25

Bonaldo RM, Grutter AS, Sazima I, Krajewski JP (2014) 24/7 service: nocturnal cleaning in a tropical Indo-Pacific reef.

Mar Biodivers:1-2

Bond WJ (1994) Keystone species. Springer Berlin Heidelberg

Bonin MC, Almany GR, Jones GP (2011) Contrasting effects of habitat loss and fragmentation on coral-associated reef

fishes. Ecology 92:1503-1512

104

Bosiger YJ, McCormick MI (2014) Temporal links in daily activity patterns between coral reef predators and their prey.

PLoS ONE 9:e111723

Bott NJ, Cribb TH (2009) Prosorhynchine trematodes (Digenea: Bucephalidae) from epinephelines (Perciformes:

Serranidae) on the Great Barrier Reef, Australia. Syst Parasitol 72:57-69

Bott NJ, Miller TL, Cribb TH (2013) Bucephalidae (Platyhelminthes: Digenea) of Plectropomus (Serranidae:

Epinephelinae) in the tropical Pacific. Parasitol Res 112:2561-2584

Bowles J, Hope M, Tiu WU, Liu X, McManus DP (1993) Nuclear and mitochondrial genetic markers highly conserved

between Chinese and Philippines Schistosoma japonicum. Acta Trop 55:217-229

Bray R, Cribb T (2002) Lepocreadiinae lepocreadiids (Digenea) from tetraodontiform fishes in the waters off Tasmania,

Queensland and Moorea, French Polynesia. Acta Parasitol 47:6-13

Bray R, Cribb T (2004) Species of Lecithocladium Luhe, 1901 (Digenea, Hemiuridae) from Australian marine fishes,

with a description of a new species from various hosts off eastern Australia. Acta Parasitol 49:3-11

Bray RA (1982) Two new species of Bacciger Nicoll, 1914 (Digenea: Fellodistomidae) from mullet in Australia. J Nat

Hist 16:23-29

Bray RA (2001) Family Tandanicolidae Johnston, 1927. In: Gibson DI, Jones A, Bray RA (eds) Keys to the Trematoda,

Volume 1. CABI Publishing and the Natural History Museum, Wallingford,

Bray RA (2002) Superfamily Gymnophalloidea Odhner, 1905. In: Gibson DI, Jones A, Bray RA (eds) Keys to the

Trematoda Vol 1. CAB International, Wallingford, pp243-244

Bray RA (2008) Family Faustulidae Poche, 1926. In: Bray RA, Gibson DI, Jones A (eds) Keys to the Trematoda Vol 3.

CABI Publishing and the Natural History Museum, Wallingford, pp509-522

Bray RA, MacKenzie K (1990) Aponurus laguncula Looss, 1907 (Digenea: Lecithasteridae): a report from herring,

Clupea harengus L., in the easter English Channel and a review of its biology. Syst Parasitol 17:115-124

Bray RA, Cribb TH (1996) Two Lepotrema Ozaki, 1932 species (Digenea: Lepocreadiidae) from marine fishes from

the southern Great Barrier Reef, Queensland, Australia. Syst Parasitol 35:111-117

Bray RA, Cribb TH (1998) Lepocreadiidae (Digenea) of Australian coastal fishes: new species of Opechona Loos,

1907, Lepotrema Ozaki, 1932 and Bianium Stunkard, 1930 and comments on other species reported for the

first time or poorly known in Australian waters. Syst Parasitol 41:123-148

Bray RA, Cribb TH, Barker SC (1993a) The Hemiuroidea (Digenea) of pomacentrid fishes (Perciformes) from Heron

Island, Queensland, Australia. Syst Parasitol 24:159-184

Bray RA, Cribb TH, Barker SC (1993b) The Lepocreadiidae (Digenea) of pomacentrid fishes (Perciformes) from Heron

Island, Queensland, Australia. Systematic Parasitology 26:189-200

Bray RA, Cribb TH, Barker SC (1994) Fellodistomidae and Lepocreadiidae (Platyhelminthes: Digenea) from

chaetodontid fishes (Perciformes) from Heron Island, southern Great Barrier Reef, Queensland, Australia.

Invertebr Taxon 8:545-581

Bray RA, Littlewood DTJ, Herniou EA, Williams B, Henderson RE (1999) Digenean parasites of deep-sea teleosts: a

review and case studies of intrageneric phylogenies. Parasitology 119:S125-S144

Bray RA, Waeschenbach A, Cribb TH, Weedall GD, Dyal P, Littlewood DTJ (2009) The phylogeny of the

Lepocreadioidea (Platyhelminthes, Digenea) inferred from nuclear and mitochondrial genes: Implications for

their systematics and evolution. Acta Parasitol 54:310-329

Bronstein JL (1994) Conditional outcomes in mutualistic interactions. Trends Ecol Evol 9:214-217

Bronstein JL, Dieckmann U, Ferrière R (2004) Evolutionary Conservation Biology. Cambridge University Press,

Cambridge

Brown BL, Creed RP, Skelton J, Rollins MA, Farrell KJ (2012) The fine line between mutualism and parasitism:

complex effects in a cleaning symbiosis demonstrated by mutliple field experiments. Oecologia 170:199-207

Brown SP, Renaud F, Guegan JF, Thomas F (2001) Evolution of trophic transmission in parasites: the need to reach a

mating place? J Evol Biol 14:815-820

Bshary R (2003) The cleaner wrasse, Labroides dimidiatus, is a key organism for reef fish diversity at Ras Mohammed

National Park, Egypt. J Anim Ecol 72:169-176

Bshary R, Würth M (2001) Cleaner fish Labroides dimidiatus manipulate client reef fish by providing tactile

stimulation. Proc R Soc Biol Sci Ser B 268:1495-1501

Bshary R, Oliveira RF, Oliveira TS, Canário AV (2007) Do cleaning organisms reduce the stress response of client reef

fish? Front Zool 4:21

Bykhovskaya-Pavlovskaya I, Gusev A, Dubinia M, Izyumova N, Smirnova T, Sokolovskaya I, Shtein G, Shul'man S,

Epshtein V (1962) Key to parasites of freshwater fish of the USSR Akad. Sci USSR Zool Inst Keys to fauna of

the USSR 80:180-223

Caley MJ, Carr MH, Hixon MA, Hughes TP, Jones GP, Menge BA (1996) Recruitment and the local dynamics of open

marine populations. Annu Rev Ecol Syst 27:477-500

Candolin U, Voigt HR (2001) No effect of a parasite on reproduction in stickleback males: a laboratory artefact?


Carrassón M, Cribb TH (2014) Benign effect of the fish parasitic isopod Ceratothoa cf. imbricata on Selenotoca

multifasciata (Scatophagidae) from Australia. Dis Aquat Organ 110:173-180

105

Carreras-Aubets M, Repullés-Albelda A, Kostadinova A, Carrassón M (2011) A new cryptic species of Aponurus

Looss, 1907 (Digenea: Lecithasteridae) from Mediterranean goatfish (Teleostei: Mullidae). Syst Parasitol

79:145-159

Ceccarelli D (2007) Modification of benthic communities by territorial damselfish: a multi-species comparison. Coral

Reefs 26:853-866

Ceccarelli D, Jones G, McCook L (2005) Effects of territorial damselfish on an algal-dominated coastal coral reef.

Coral Reefs 24:606-620

Chambers JM (1992) Linear models. In: Chambers JM, Hastie TJ (eds) Statistical models in S. Wadsworth &

Brooks/Cole, Pacific Grove, California, pp95-144

Cheney KL, Côté IM (2001) Are Caribbean cleaning symbioses mutualistic? Costs and benefits of visiting cleaning

stations to longfin damselfish. Anim Behav 62:927-933

Cheney KL, Côté IM (2003) The ultimate effect of being cleaned: does ectoparasite removal have reproductive

consequences for damselfish clients? Behav Ecol 14:892-896

Cheney KL, Bshary R, Grutter AS (2008) Cleaner fish cause predators to reduce aggression toward bystanders at

cleaning stations. Behav Ecol 19:1063-1067

Cheney KL, Grutter AS, Blomberg SP, Marshall NJ (2009) Blue and yellow signal cleaning behavior in coral reef

fishes. Curr Biol 19:1283-1287

Chun SK, Chang DS, Park CK, Kim YG, Rho YG (1981) Basic studies for the production of the hard clam Meretrix

lusoria (Roding) in Jeonnbug farming area. Bulletin of Fisheries Research and Development Agency 26:7-36

Clague GE, Cheney KL, Goldizen AW, McCormick MI, Waldie PA, Grutter AS (2011) Long-term cleaner fish

presence affects growth of a coral reef fish. Biol Lett 6:863-865

Coile AM, Sikkel PC (2013) An experimental field test of susceptibility to ectoparasitic gnathiid isopods among

Caribbean reef fishes. Parasitology 140:888-896

Coker DJ, Pratchett MS, Munday PL (2012a) Influence of coral bleaching, coral mortality and conspecific aggression

on movement and distribution of coral-dwelling fish. J Exp Mar Biol Ecol 414-415:62-68

Coker DJ, Graham NAJ, Pratchett MS (2012b) Interactive effects of live coral and structural complexity on the

recruitment of reef fishes. Coral Reefs 31:919-927

Cooper WJ, Smith JJ, Westneat MW (2009) Exploring the radiation of a diverse reef fish family: phylogenetics of the

damselfishes (Pomacentridae), with new classifications based on molecular analyses of all genera. Mol

Phylogenet Evol 52:1-16

Coppock AG, Gardiner NM, Jones GP (2013) Olfactory discrimination in juvenile coral reef fishes: Responses to

conspecifics and corals. J Exp Mar Biol Ecol 443:21-26

Côté IM (2000) Evolution and ecology of cleaning symbioses in the sea. Oceanography and Marine Biology 38:311-

355

Côté IM, Arnal C, Reynolds JD (1998) Variation in posing behaviour among fish species visiting cleaning stations. J

Fish Biol 53 (sA):256-266

Cribb T, Bray R (1994) A review of the Tandanicolinae (= Monodhelminthinae)(Digenea: Fellodistomidae), with a

description of Prosogonarium angelae n. sp. Syst Parasitol 27:187-205

Cribb TH, Bray RA (2010) Gut wash, body soak, blender and heat-fixation: approaches to the effective collection,

fixation and preservation of trematodes of fishes. Syst Parasitol 76:1-7

Cribb TH, Bray RA, Barker SC (1992) Zoogonidae (Digenea) from southern Great Barrier Reef fishes with a

description of Steganoderma (Lecithostaphylus) gibsoni n. sp. Syst Parasitol 23:7-12

Cribb TH, Bray RA, Barker SC (1994a) Bivesiculidae and Haplosplanchnidae (Digenea) from fishes of the southern

Great Barrier Reef, Australia. Syst Parasitol 28:81-97

Cribb TH, Anderson GR, Bray RA (1999) Faustulid trematodes (Digenea) from marine fishes of Australia. Syst

Parasitol 44:119-138

Cribb TH, Anderson GR, Adlard RD, Bray RA (1998) A DNA-based demonstration of a three-host life cycle for the

Bivesiculidae (Platyhelminthes: Digenea). Int J Parasitol 28:1791-1795

Cribb TH, Miller TL, Bray RA, Cutmore S, C. (2014a) The sexual adult of Cercaria praecox Walker, 1971 (Digenea:

Fellodistomidae), with the proposal of Oceroma n. g. Syst Parasitol 88:1-10

Cribb TH, Miller TL, Bray RA, Cutmore SC (in press) The sexual adult of Cercaria praecox Walker, 1971 (Digenea:

Fellodistomidae), with the proposal of Oceroma n. g. Syst Parasitol

Cribb TH, Bray RA, Barker SC, Adlard RD, Anderson GR (1994b) Ecology and diversity of digenean trematodes of

reef and inshore fishes of Queensland. Int J Parasitol 24:851- 860

Cribb TH, Bray RA, Littlewood DTJ, Pichelin SP, Herniou EA (2001) The Digenea. In: Littlewood DTJ, Bray RA (eds)

Interrelationships of the Platyhelminthes. Taylor and Francis, London, pp168-185

Cribb TH, Bott NJ, Bray RA, McNamara MKA, Miller TL, Nolan MJ, Cutmore SC (2014b) Trematodes of the Great

Barrier Reef, Australia: emerging patterns of diversity and richness in coral reef fishes. Int J Parasitol 44:929-

939

Curran SS, Tkach VV, Overstreet RM (2006) A review of Polylekithum Arnold, 1934 and its familial affinities using

morphological and molecular data, with description of Polylekithum catahoulensis sp. nov. Acta Parasitol

51:238-248

106

Curtis LM, Grutter AS, Smit NJ, Davies AJ (2013) Gnathia aureamaculosa, a likely definitive host of Haemogregarina

balistapi and potential vector for Haemogregarina bigemina between fishes of the Great Barrer Reef,

Australia. Int J Parasitol 43:361-370

Danilowicz BS (1996) Choice of coral species by naive and field caught damselfishes. Copeia 3:735-739

Davies AJ, Smit NJ (2001) The life cycle of Haemogregarina bigemina (Adeleina: Haemogregarinidae) in South

African hosts. Folia Parasitol (Praha) 48:169-177

Davies AJ, Curtis L, Grutter AS, Smit NJ (2009) Suspected viral erythrocytic necrosis (VEN) in a juvenile blackbar

triggerfish, Rhinecanthus aculeatus, from Lizard Island, Great Barrier Reef, Australia. Marine Biodiversity

Records 2:e149

Davies AJ, Smit NJ, Hayes PM, Seddon AM, Wertheim D (2004) Haemogregarina bigemina (Protozoa: Apicomplexa:

Adeleorina) - Past, present and future. Folia Parasitol (Praha) 51:99-108

Derbel H, Chaari M, Neifar L (2011) Redescription of Rhipidocotyle galeata (Rudolphi, 1819) (Digenea,

Bucephalidae), the type species of Rhipidocotyle Diesing, 1907. Zoosystema 33:133-139

Diaz EP, Bray RA, Cribb TH (2013) Paradiscogaster flindersi and P. oxleyi n. sp. (Digenea: Faustulidae): overlapping

host and geographical distributions in corallivore chaetodontid fishes in the tropical Indo-west Pacific. Syst

Parasitol 86:87-99

Diaz PE, Cribb TH (2013) Paradiscogaster leichhardti sp. nov. (Digenea: Faustulidae) in Chaetodontoplus meredithi

(Perciformes: Pomacanthidae) from Heron Island, Great Barrier Reef. Memoirs of the Queensland Museum

58:49-53

Diaz PE, Bray RA, Cutmore SC, Ward S, Cribb TH (2015) A complex of species related to Paradiscogaster glebulae

(Digenea: Faustulidae) in chaetodontid fishes (Teleostei: Perciformes) of the Great Barrier Reef. Parasitol Int

Dimitrov GI, Bray RA, Gibson DI (1999) A redescription of Pseudobacciger harengulae (Yamaguti, 1938) (Digenea:

Faustulidae) from Sprattus sprattus phalericus (Risso) and Engraulis encrasicholus ponticus Alexandrov off

the Bulgarian Black Sea coast, with a review of the genus Pseudobacciger Nahhas & Cable, 1964. Syst


Dyer WG, Williams EH, Williams LB (1985) Digenetic trematodes of marine fishes of the western and southern coasts

of Puerto Rico. Proc Helminthol Soc Wash 52:85-94

Dyer WG, Williams EH, Bunkley-Williams L (1988) Digenetic trematodes of marine fishes of Okinawa, Japan. J


Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res

32:1792-1797

Euzet L, Combes Cle (1980) Les problèmes de l'espèce chez les animaux parasites. Bull Soc Zool Fr 40:239-285

Feary DA, Almany GR, McCormick MI, Jones GP (2007) Habitat choice, recruitment and the response of coral reef

fishes to coral degradation. Oecologia 153:727-737

Ferguson JA, Koketsu W, Ninomiya I, Rossignol PA, Jacobson KC, Kent ML (2011) Mortality of coho salmon

(Oncorhynchus kisutch) associated with burdens of multiple parasite species. Int J Parasitol 41:1197-1205

Finley RJ, Forrester GE (2003) Impact of ectoparasites on the demography of a small reef fish. Mar Ecol Prog Ser

248:305-309

Fishelson L (1998) Behaviour, socio-ecology and sexuality in damselfishes (Pomacentridae). Ital J Zool 65:387-398

Fogelman RM, Grutter AS (2008) Mancae of the parasitic cymothoid isopod, Anilocra apogonae: early life history,

host-specificty, and effect on growth and survival of preferred young cardinal fishes. Coral Reefs 27:685-693

Fogelman RM, Kuris AM, Grutter AS (2009) Parasitic castration of a vertebrate: effect of the cymothoid isopod,

Anilocra apogonae, on the five-lined cardinalfish, Cheilodipterus quinquelineatus. Int J Parasitol 39:577-583

Froese R, Pauly D (2012) FishBase (version 02/2014), www.fishbase.org

Gaevskaja AV (1996) [New records of trematodes from eastern Atlantic fishes] (In Russian). Parasitlogiya 30:504-509

Gaevskaja AV, Naidenova NN (1996) [Revision of Bacciger bacciger (Trematoda: Fellodistomidae) from fishes of the

Black and Mediterranean Seas] (In Russian). Parazitologiya (St Petersb) 30:39-44

Gibson DI, Bray RA (1979) The Hemiuroidea: terminology, systematics and evolution. Bulletin of the British Museum

Natural History (Zoology) 36:35-146

Gorlick DL, Atkins PD, Losey GS (1987) Effect of cleaning by Labroides dimidiatus (Labridae) on an ectoparasite

population infecting Pomacentrus vaiuli (Pomacentridae) at Enewetak Atoll. Copeia 1:41-45

Green AL (1994) The early life history of labroid fishes at Lizard Island, Northern Great Barrier Reef. James Cook

University,

Green AL (1998) Spatio-temporal patterns of recruitment of labroid fishes (Pisces: Labridae and Scaridae) to

damselfish territories. Environ Biol Fishes 51:235-244

Green BS (2008) Maternal effects in fish populations. Adv Mar Biol 54:1-105

Grorud-Colvert K, Sponaugle S (2011) Variability in water temperature affects trait-mediated survival of a newly

settled coral reef fish. Oecologia 165:675-686

Grutter AS (1994) Spatial and temporal variations of the ectoparasites of seven reef fish species from Lizard Island and

Heron Island, Australia. Mar Ecol Prog Ser 115:21-30

Grutter AS (1995a) Relationship between cleaning rates and ectoparasite loads in coral reef fishes. Mar Ecol Prog Ser

118:51-58

http://www.fishbase.org/

107

Grutter AS (1995b) A comparison of methods for sampling ectoparasites from coral reef fishes. Mar Freshw Res

46:897-903

Grutter AS (1996a) Parasite removal rates by the cleaner wrasse, Labroides dimidiatus. Mar Ecol Prog Ser 130:61-70

Grutter AS (1996b) Experimental demonstration of no effect by the cleaner Labroides dimidiatus (Cuvier and

Valenciennes) on the host fish Pomacentrus moluccensis (Bleeker). J Exp Mar Biol Ecol 196:285-298

Grutter AS (1997) Spatiotemporal variation and feeding selectivity in the diet of the cleaner fish Labroides dimidiatus.

Copeia 1997:346-355

Grutter AS (1999a) Cleaner fish really do clean. Nature 398:672-673

Grutter AS (1999b) Infestation dynamics of gnathiid isopod juveniles parasitic on the coral-reef fish Hemigymnus

melapterus (Labridae). Mar Biol 135:545-552

Grutter AS (2000) Ontogenetic variation in the diet of the cleaner fish Labroides dimidiatus and its ecological

consequences Mar Ecol Prog Ser 197:241-246

Grutter AS (2001) Parasite infection rather than tactile stimulation is the proximate cause of cleaning behaviour in reef

fish. Proc R Soc Biol Sci Ser B 268:1361-1365

Grutter AS (2002) Cleaning symbioses from the parasites' perspective. Parasitology 124:65-81

Grutter AS (2003) Feeding ecology of the fish ectoparasite Gnathia sp. (Crustacea: Isopoda) from the Great Barrier

Reef, and its implications for cleaning behaviour. Mar Ecol Prog Ser 2559:295-302

Grutter AS (2004a) Cleaner fish use tactile dancing behaviour as a preconflict management strategy. Curr Biol 14:1080-

1083

Grutter AS (2004b) Cleaner fish use tactile dancing behaviour as a preconflict management strategy. Current biology

14:1080-1083

Grutter AS (2008) Interactions between gnathiid isopods, cleaner fish and other fishes on Lizard Island, Great Barrier

Reef. J Fish Biol 73:2094-2109

Grutter AS, Poulin R (1998) Intraspecific and interspecific relationships between host size and the abundance of

parasitic larval gnathiid isopods on coral reef fishes. Mar Ecol Prog Ser 164:263-271

Grutter AS, Hendrikz J (1999) Diurnal variation in the abundance of juvenile parasitic gnathiid isopods on coral reef

fish- implications for parasite-cleaner fish interactions. Coral Reefs 18:187-191

Grutter AS, Lester RJG (2002) Cleaner fish Labroides dimidiatus reduce 'temporary' parasitic corallanid isopods on the

coral reef fish Hemigymnus melapterus. Mar Ecol Prog Ser 234:247-255

Grutter AS, Lester RJG, Greenwood JG (2000) Emergence rates from the benthos of the parasitic juveniles of gnathiid

isopods. Mar Ecol Prog Ser 207:123-127

Grutter AS, Murphy J, Choat H (2003) Cleaner fish drives local fish diversity on coral reefs. Curr Biol 13:64-67

Grutter AS, Glover S, Bshary R (2005) Does client size affect cleaner fish choice of client? An empirical test using

client fish models. J Fish Biol 66:1748-1752

Grutter AS, Pickering JL, McCallum H, McCormick MI (2008) Impact of micropredatory gnathiid isopods on young

coral reef fishes. Coral Reefs 27:655-661

Grutter AS, Cribb TH, McCallum H, Pickering JL, McCormick MI (2010) Effects of parasites on larval and juvenile

stages of the coral reef fish Pomacentrus moluccensis. Coral Reefs 29:31-40

Grutter AS, Crean AJ, Curtis LM, Kuris AM, Warner RR, McCormick MI (2011) Indirect effect of an ectoparasite

reduce successful establishment of a damselfish at settlement. Funct Ecol 25:586-594

Grutter AS, Blomberg SP, Fargher B, Kuris AM, McCormick MI, Warner RR (in review) A reason for pelagic larval

stages: Small size of coral reef fish at settlement and high mortality risk from gnathiid isopod micropredators

Guégan JF, Lambert A, Lévêque C, Combes C, Euzet L (1992) Can host body size explain the parasite species richness

in tropical freshwater fishes? Oecologia 90

Hall KA, Cribb TH, Barker SC (1999) V4 region of small subunit rDNA inidicates polyphyly of the Fellodistomidae

(Digenea) which is supported by morphology and life-cycle data. Syst Parasitol 43:81-92

Harms KE, Wright SJ, Calderón O, Hernández A, Herre EA (2000) Pervasive density-dependent recruitment enhances

seedling diveristy in a tropical forest. Nature 404:493-495

Hayes PM, Smit NJ, Grutter AS, Davies AJ (2011) Unexpected response of a captive blackeye thicklip, Hemigymnus

melapterus (Bloch), from Lizard Island, Australia, exposed to juvenile isopods Gnathia aureamaculosa

Ferreira & Smit. J Fish Dis 34:563-566

Heinlein J, M., Stier AC, Steele MA (2010) Predators reduce abundance and species richness of coral reef fish recruits

via non-selective predation. Coral Reefs 29:527-532

Helfman GS (1986) Fish behaviour by day, night and twilight The behaviour of teleost fishes. Springer, pp366-387

Heupel MR, Bennett MB (1999) The occurrence, distribution and pathology associated with gnathiid isopod larvae

infecting the epaulette shark, Hemiscyllium ocellatum. International Journal for Parasitology 29:321-330

Hixon MA, Brostoff WN (1983) Damselfish as keystone species in reverse: intermediate disturbance and diversity of

reef algae. Science 220:511-513

Ho HW, Bray RA, Cutmore SC, Ward S, Cribb TH (2014) Two new species of Phyllodistomum Braun, 1899

(Trematoda: Gorgoderidae Looss, 1899) from Great Barrier Reef fishes. Zootaxa 3779:551-562

Hoey AS, McCormick MI (2004) Selective predation for low body condition at the larval-juvenile transition of a coral

reef fish. Oecologia 139:23-29

108

Horthorn T, Bretz F, Westfall P, Hothorn MT (2013) Package 'multcomp' (ed. CRAN) See http://cran.r-

project.org/web/packages/multcomp.

Hudson PJ, Dobson AP, Lafferty KD (2006) Is a healthy ecosystem one that is rich in parasites? Trends Ecol Evol

21:381-385

Huijbers CM, Nagelkerken I, Lössbroek PA, Schulten IE, Siegenthaler A, Holderied MW, Simpson SD (2012) A test of

the senses: Fish select novel habitats by responding to multiple cues. Ecology 93:46-55

Hunter JA, Cribb TH (2010) Transversotrematidae (Plathelminthes: Trematoda) are rich and abundant on Indo-Pacific

fishes. Zootaxa 2442:25-38

Hunter JA, Cribb TH (2012) A cryptic complex of species related to Transversotrema licinum Manter, 1970 from fishes

of the Indo-West Pacific, including descriptions of ten new species of Transversotrema Witenberg, 1944

(Digenea: Transversotrematidae). Zootaxa 3176:1-44

Ichihara A (1974) Hemiurid trematodes from marine fishes near the Tsushima Islands in the Sea of Japan. Third

International Congress of Parasitology

Ichihara A (1986) On the helminths of marine fishes in Sagami Bay (II) Parasitology - quo vadit? Sixth International

Congress of Parasitology

Jacoby CA, Greenwood JG (1988) Spatial, temporal, and behavioural patterns in emergence of zooplankton in the

lagoon of Heron Reef, Great Barrier Reef, Australia. Mar Biol 97:309-328

Johnston SJ (1913) On some Queensland trematodes, with anatomical observations and descriptions of new species and

genera. Q J Microsc Sci 59:361-400

Johnston TH (1927) New trematodes from an Australian siluroid. Trans R Soc S Aust 51:129-136

Jones CM, Grutter AS (2005) Parasitic isopods (Gnathia sp.) reduce haematocrit in captive blackeye thicklip (Labridae)

on the Great Barrier Reef. Journal of Fish Biology 66:860-864

Jones CM, Grutter AS (2007) Variation in emergence of parasitic and predatory isopods among habitats at Lizard

Island, Great Barrier Reef. Mar Biol 150:919-927

Jones CM, Grutter AS (2008) Reef-based micropredators reduce the growth of post-settlement damselfish in captivity.

Coral Reefs 27:677-684

Jones GP (1988) Experimental evaluation of the effects of habitat structure and competitive interactions on the juveniles

of two coral reef fishes. J Exp Mar Biol Ecol 123:115-126

Jordan LKB, Lindeman KC, Spieler RE (2012) Depth-variable settlement patterns and predation influence on newly

settled reef fish (Haemulon spp., Haemulidae). PLoS ONE 7:e50897

Jordano P (2000) Fruits and frugivory. In: Fenner M (ed) Seeds: The ecology of regeneration in natural plant

communities. CABI, Wallingford, UK, pp125-166

Jousson O, Bartoli P, Pawlowski J (2000) Cryptic speciation among intestinal parasites (Trematoda: Digenea) infecting

sympatric host fishes (Sparidae). J Evol Biol 13:778-785

Justine JL (2010) Parasites of coral reef fish: how much do we know? With a bibliography of fish parasites in New

Caledonia. Belg J Zool 140:155-190

Justine JL, Beveridge I, Boxshall GA, Bray RA, Moravec F, Trilles JP, Whittington ID (2010) An annotated list of

parasites (Isopoda, Copepoda, Monogenea, Digenea, Cestoda and Nematoda) collected in groupers

(Serranidae, Epinephelinae) in New Caledonia emphasizes parasite biodiversity in coral reef fish. Folia

Parasitol (Praha) 57:237-262

Kamiya T, O'Dwyer K, Nakagawa S, Poulin R (2014) What determines species richness of parasitic organisms? A

meta‐analysis across animal, plant and fungal hosts. Biological Reviews 89:123-134

Kerrigan BA (1996) Temporal patterns in size and condition at settlement in two tropical reef fishes (Pomacentridae:

Pomacentrus amboinensis and P. nagasakiensis). Mar Ecol Prog Ser 135:27-41

Kim YG, Chun SK (1984) Studies on the life history of Bacciger harengulae. J Korean Fish Soc 17:449-470

King RE (1964) Three hemiurid trematodes from South Vietnam. Trans Am Microsc Soc 83:435-439

Knowlton N, Rohwer F (2003) Multispecies microbial mutualisms on coral reefs: The host as a habitat. The American

Naturalist 162:51-62

Korotaeva VD (1969) [Helminths of some food-fishes of the order Clupeiformes in the sea near Australia and New

Zealand] (In Russian). In: Markevich AP (ed) Problemy Parazitologii Trudy Nauchno Konferentsii

Parazitologov USSR (6th) 237-238

Krebs CJ (1999) Ecological methodology. Benjamin/Cummings, Menlo Park, California

Krkošek M, Revie CW, Gargan PG, Skilbrei OT, Finstad B, Todd CD (2013) Impact of parasites on salmon recruitment

in the Northeast Atlantic Ocean. Proc R Soc Biol Sci Ser B 280:doi: 10.1098/rspb.2012.2359

Kuris A, Lafferty K (2000) Parasite-host modeling meets reality: adaptive peaks and their ecological attributes.

Evolutionary biology of host–parasite relationships: theory meets reality:9-26

Kurochkin YV (1970) [New trematode species Discogasteroidinae (Trematoda, Fellodistomatidae) from marine fishes

of Australia] (In Russian). Parazitologiya 4:111-115

Lafferty KD, Kuris AM (2002) Trophic strategies, animal diversity and body size. Trends Ecol Evol 17:507-513

Lecchini D, Planes S, Galzin R (2005) Experimental assessment of sensory modalities of coral-reef fish larvae in the

recognition of their settlement habitat. Behav Ecol Sociobiol 58:18-26

Lecchini D, Planes S, Galzin R (2007a) The influence of habitat characteristics and conspecifics on attration and

survival of coral reef fish juveniles. J Exp Mar Biol Ecol 341:85-90

http://cran.r-project.org/web/packages/multcomp

http://cran.r-project.org/web/packages/multcomp

109

Lecchini D, Peyrusse K, Lanyon RG, Lecellier G (2014) Importance of visual cues of conspecifics and predators during

the habitat selection of coral reef fish larvae. Ethology 337:345-351

Lecchini D, Osenberg CW, Shima JS, St Mary CM, Galzin R (2007b) Ontogenetic changes in habitat selection during

settlement in a coral reef fish: ecological determinants and sensory mechanisms. Coral Reefs 26:423-432

Lemly AD, Esch GW (1984) Effects of the trematode Uvulifer ambloplitis on juvenile Bluegill Sunfish, Lepomis

macrochirus: ecological implications. J Parasitol 70:475-492

Lester RJG, Sewell KB (1990) Checklist of parasites from Heron Island, Great Barrier Reef. Aust J Zool 37:101-128

Lewis AR (1997) Recruitment and post-recruit immigration affect the local population size of coral reef fishes. Coral

Reefs 16:139-149

Lienart GDH, Mitchell MD, Ferrari MC, McCormick MI (2014) Temperature and food availability affect risk

assessment in an ectotherm. Anim Behav 89:199-204

Linton E (1910) Helminth fauna of the Dry Tortugas. II. Trematodes. Papers of the Tortugas Laboratory of the Carnegie

Institute of Washington Publication 4:11-98

Littlewood DTJ, Curini-Galletti M, Herniou EA (2000) The interrelationships of Proseriata (Platyhelminthes: Seriata)

tested with molecules and morphology. Mol Phylogenet Evol 16:449-466

Lo CM, Morand S, Galzin R (1998) Parasite diversity/host age and size relationship in three coral-reef fishes from

French Polynesia. Int J Parasitol 28:1695-1708

Losey GS (1972) The ecological importance of cleaning symbiosis. Copeia 4:820-833

Losey GS (1979) Fish cleaning symbiosis: proximate cauases of host behaviour. Anim Behav 27:669-685

Losey GS, Mahon JL, Danilowicz BS (1995) Innate recognition by host fish of their cleaning symbiont. Ethology

100:277-283

Machida M (1984) Two new trematodes from gallbladder of tropical marine fishes, Myripristis and Abudefduf. Bulletin

of the Natural Museum, Tokyo 10:1-5

Machida M (2005) Monorchiidae (Trematoda, Digenea) from fishes of Japanese and adjacent waters. Bulletin of the

National Science Museum, Tokyo Series A Zoology 31:123-136

Machida M, Uchida K (2001) Digenean trematodes from pomacenthid fishes of Japan and Palau. Bulletin of the

National Science Museum, Tokyo, Series A, Zoology 27:217-227

Maddison WP, Maddison DR (2009) Mesquite: a modular system for evolutionary analysis. Version 2.72

http://mesquiteproject.org

Madhavi R (1975) Digenetic trematodes from marine fishes of Waltair coast, Bay of Bengal: Family Fellodistomidae.

Revista di Parasitologia 36:115-128

Manter HW (1934) Some digenetic trematodes from deep-water fish of Tortugas, Florida. Papers from Tortugas

Laboratory 28:257-345

Manter HW (1942) Monorchiidae (Trematoda) from fishes of Tortugas, Florida. Trans Am Microsc Soc 61:349-360

Manter HW (1947) The digenetic trematodes of marine fishes of Tortugas, Florida. The American Midland Naturalist

38:257-416

Manter HW (1969a) Three new genera hemiurid trematodes from Australian fishes. Journal of the Fisheries Board of

Canada 26:787-792

Manter HW (1969b) Some digenetic trematodes of marine fishes of New Caledonia. Part IV. Hemiuridae and summary.

Proc Helminthol Soc Wash 36:194-204

Marcogliese DJ (1995) The role of zooplankton in the transmission of helminth parasites to fish. Rev Fish Biol Fish

5:336-371

Marcogliese DJ (2002) Food webs and the transmission of parasites to marine fish. Parasitology 124:83-99

Marino F, Giannetto S, Paradiso ML, Bottari T, De Vico G, Macri B (2004) Tissue damage and haematophagia due to

praniza larvae (Isopoda: Gnathiidae) in some aquarium seawater teleosts. Dis Aquat Organ 59:43-47

McCormick MI (2006) Mothers matter: crowding leads to stressed mothers and smaller offspring in marine fish.

Ecology 87:1104-1109

McCormick MI, Hoey AS (2004) Larval growth history determines juvenile growth and survival in a tropical marine

fish. Oikos 106:225-242

McCormick MI, Meekan MG (2007) Social facilitation of selective mortality Ecology 88:1562-1570

McCormick MI, Moore JAY, Munday PL (2010) Influence of habitat degradation on fish replenishment. Coral Reefs

29:537-546

McNamara M, Cribb T (2011) Taxonomy, host specificity and dietary implications of Hurleytrematoides (Digenea:

Monorchiidae) from chaetodontid fishes on the Great Barrier Reef. Parasitol Int 60:255-269

McNamara M, Miller T, Cribb T (2014) Evidence for extensive cryptic speciation in trematodes of butterflyfishes

(Chaetodontidae) of the tropical Indo-West Pacific. Int J Parasitol 44:37-48

McNamara M, Adlard R, Bray R, Sasal P, Cribb T (2012) Monorchiids (Platyhelminthes: Digenea) of chaetodontid

fishes (Perciformes): biogeographical patterns in the tropical Indo-West Pacific. Parasitol Int 61:288-306

Medeiros PR, Souza AT, Ilarri MI (2010) Habitat use and behavioural ecology of the juveniles of two sympatric

damselfishes (Actinopterygii: Pomacentridae) in the south-western Atlantic Ocean. J Fish Biol 77:1599-1615

Meekan M, Steven A, Fortin M (1995) Spatial patterns in the distribution of damselfishes on a fringing coral reef. Coral

Reefs 14:151-161

http://mesquiteproject.org/

110

Meekan MG, Wilson SG, Halford A, Retzel A (2001) A comparison of catches of fishes and invertebrates by two light

trap designs, in tropical NW Australia. Mar Biol 139:373-381

Mensink PJ, Geange SW, Shima JS (2014) Reproductive success of parasitized males in a marine reef fish. Mar Biol

161:2689-2696

Milicich MJ, Doherty PJ (1994) Larval supply of coral reef fish populations - magnitude and synchrony of

replenishment to Lizard Island, Great Barrier Reef. Mar Ecol Prog Ser 110:121-134

Militz TA, Hutson KS (2015) Beyond symbiosis: cleaner shrimp clean up in culture. PLoS ONE 10:e0117723

Miller MA, Pfeiler E, Schwartz T (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic

trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov. 2010, New

Orleans, LA, pp. 1-8

Miller TL, Cribb TH (2007) Coevolution of Retrovarium n. gen. (Digenea: Cryptogonimidae) in Lutjanidae and

Haemulidae (Perciformes) in the Indo-West Pacific. Int J Parasitol 37:1023-1045

Miller TL, Cribb TH (2008) Eight new species of Siphoderina Manter, 1934 (Digenea, Cryptogonimidae) infecting

Lutjanidae and Haemulidae (Perciformes) off Australia. Acta Parasitol 53:344-364

Miller TL, Bray RA, Cribb TH (2011a) Taxonomic approaches to and interpretation of host-specificity of trematodes of

fishes: lessons from the Great Barrier Reef. Parasitology 138:1710-1722

Miller TL, Bray RA, Cribb TH (2011b) Taxonomic approaches to and interpretation of host specificity of trematodes of

fishes: lessons from the Great Barrier Reef. Parasitology

Monod T (1926) Les Gnathiidae. Essai monographique (morphologie, biologie, systématique). Mémoires de la Société

des Sciences Naturelles du Maroc 13:1-668

Morand S (2015) (macro-) Evolutionary ecology of parasite diversity: From determinants of parasite species richness to

host diversification. International Journal for Parasitology: Parasites and Wildlife 4:80-87

Morand S, Cribb TH, Kulbicki M, Rigby MC, Chauvet C, Dufour V, Faliex E, Galzin R, Lo CM, Lo-Yat A, Pichelin S,

Sasal P (2000) Endoparasite species richness of New Caledonian butterfly fishes: host density and diet matter.


Mugridge RER, Stallybrass HG (1983) A mortality of eels, Anguilla anguilla L., attributed to Gnathiidae. J Fish Dis

6:81-82

Muñoz G, Cribb TH (2006) Parasite communities and diet of Coris batuensis (Pisces: Labridae) from Lizard Island,

Great Barrier Reef. Mem Queensl Mus 52:191-198

Muñoz G, Grutter A, Cribb T (2006) Endoparasite communities of five fish species (Labridae: Cheilininae) from Lizard

Island: how important is the ecology and phylogeny of the hosts? Parasitology 132:363-374

Muñoz G, Grutter AS, Cribb TH (2007) Structure of the parasite communities of a coral reef fish assemblage

(Labridae): testing ecological and phylogenetic host factors. J Parasitol 93:17-30

Nahhas FM, Cable RM (1964) Digenetic and aspidogasterid trematodes from marine fishes of Curacao and Jamaica.

Tulane Studies of Zoology 11:167-228

Nahhas FM, Wetzel JA (1995) Digenetic trematodes of marine fishes from Suva, Fiji: the family Gyliauchenidae Ozaki,

1933. Journal of Helminthological Society Washington 62:117-130

Nolan MJ, Cribb TH (2005) The use and implications of ribosomal DNA sequencing for the discrimination of digenean

species. Adv Parasitol 60:101-163

Ohman MC, Munday PL, Jones GP, Caley MJ (1998) Settlement strategies and distribution patterns of coral-reef fishes.

J Exp Mar Biol Ecol 225:219-238

Olson PD, Cribb TH, Tkach VV, Bray RA, Littlewood DTJ (2003) Phylogeny and classification of the Digenea

(Platyhelminthes: Trematoda). Int J Parasitol 33:733-755

Ormond RFG, Roberts JM, Jan R-Q (1996) Behavioural differences in microhabitat use by damselfishes

(Pomacentridae): implications for reef fish biodiversity. J Exp Mar Biol Ecol 202:85-95

Ostlund-Nilsson S, Curtis L, Nilsson G, Grutter AS (2005) Parasitic isopod Anilocra apogonae, a drag for the cardinal

fish Cheilodipterus quinquelineatus. Mar Ecol Prog Ser 287:209-216

Ota Y, Hoshino O, Hirose M, Tanaka K, Hirose E (2012) Third-stage larva shifts host fish from teleost to elasmobranch

in the temporary parasitic isopod, Gnathia trimaculata (Crustacea: Gnathiidae). Mar Biol 159:2333-2347

Overstreet RM (1969) Digenetic trematodes from Biscayne Bay, Florida. Tulane Stud Zool Bot 15:119-176

Ozaki Y (1932) A new trematode from File-fish: Lepotrema clavatum ngn sp. Proceedings of the Imperial Academy

8:44-47

Palacios-Fuentes P, Landaeta MF, Munoz G, Plaza G, Ojeda FP (2012) The effects of a parasitic copepod on the recent

larval growth of a fish inhabiting rocky coasts. Parasitol Res 111:1661-1671

Pankov P, Webster BL, Blasco-Costa I, Gibson DI, Littlewood DTJ, Balbuena JA, Kostadinova A (2006) Robinia

aurata n. g., n. sp. (Digenea: Hemiuridae) from the mugilid Liza aurata with a molecular confirmation of its

position within the Hemiuroidea. Parasitology 133:217-227

Paperna I (1977) Preliminary data on the Gnathiidae (Isopoda) of the northern Red Sea, the Bitter Lakes and the Eastern

Mediterranean and the biology of Gnathia piscivora n. sp. Rapports et Proces-Verbaux des Reunions-

Commission Internationale pour l'Exploration Scientifique de la Mer Mediterranee (CIESM)

Paris CB, Atema J, Irisson J-O, Kingsford M, Gerlach G, Guigand CM (2013) Reef odor: a wake up call for navigation

in reef fish larvae. PLoS ONE 8:e72808

Parukhin AM (1976a) [Parasitic worms of benthic fishes of the southern seas] (In Russian). Naukova Dumka, Kiev

111

Parukhin AM (1976b) [Parasitic worms of commercially important fishes of the southern Seas] (In Russian), Kiev:

Naukova Dumka

Parukhin AM (1989) [Parasitic worms of bottom fishes of the southern seas] (In Russian). Naukova Dumka, Kiev 153

Pearson JC (1968) Observations on the morphology and life-cycle of Paucivitellosus fragilis Coil, Reid & Kuntz, 1965

(Trematoda: Bivesiculidae). Parasitology 58:769-788

Penfold R, Grutter AS, Kuris AM, McCormick MI, Jones CM (2008) Interactions between juvenile marine fish and

gnathiid isopods: predation versus micropredation. Mar Ecol Prog Ser 357:111-119

Pleijel F, Jondelius U, Norlinder E, Nygren A, Oxelman B, Schander C, Sundberg P, Thollesson M (2008) Phylogenies

without roots? A plea for the use of vouchers in molecular phylogenetic studies. Mol Phylogenet Evol 48:369-

371

Posada D (2008) jModelTest: Phylogenetic model averaging. Mol Biol Evol 25:1253-1256

Poulin R, Grutter AS (1996) Cleaning symbioses: Proximate and adaptive explanations. Bioscience 46:512-517

Poulin R, Morand S (2000) The diversity of parasites. Q Rev Biol 75:277-293

Poulin R, Cribb TH (2002) Trematode life cycles: short is sweet? Trends Parasitol 18:176-183

Pritchard MH (1963) Studies on digenetic trematodes of Hawaiian fishes, primarily families Lepocreadiidae and

Zoogonidae. The Journal of Parasitology 49:578-587

Pritchard MH, Manter HW (1961) Studies of digenetic trematodes of Hawaiian fishes-family Haplosplanchnidae. Proc

Helminthol Soc Wash 28:191

R Development Core Team (2013) R: A language and environment for statistical computing , R Foundation for

Statistical Computing. http://www.R-project.org), Vienna

Radford C, Stanley J, Simpson S, Jeffs A (2011) Juvenile coral reef fish use sound to locate habitats. Coral Reefs

30:295-305

Randall JE, Allen GR, Steene RC (1997) Fishes of the Great Barrier Reef and Coral Sea. Crawford House Publishing,

Bathurst, N.S.W

Rankin TL, Sponaugle S (2014) Characteristics of settling coral Rref fish are related to recruitment timing and success.

PLoS ONE 9:e108871

Rizk MH, Taha SA, Ab El-Rahman A (1996) Helminth parasites of some fishes from Mediterranean Sea and their

effect on the coefficient of condition. J Egypt Ger Soc Zool 19:129-144

Robertson DR (1974) A study of the biology and reproductive biology of the labrid fish, Labroides dimidiatus, at Heron

Island, Great Barrier Reef. Ph.D, thesis. The University of Queensland, pp 174

Roche DG, Strong LE, Binning SA (2013a) Prevalence of the parasitic cymothoid isopod Anilocra nemipteri on its fish

host at Lizard Island, Great Barrier Reef. Aust J Zool 60:330-333

Roche DG, Binning SA, Strong LE, Davies JN, Jennions MD (2013b) Increased behavioural lateralization in parasitized

coral reef fish. Behav Ecol Sociobiol 67:1339-1344

Rohde K (2002) Ecology and biogeography of marine parasites. Adv Mar Biol 43:1-86

Rohner CA, Cribb TH (2013) Opecoelidae (Digenea) in northern Great Barrier Reef goatfishes (Perciformes: Mullidae).

Syst Parasitol 84:237-253

Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Hohna S, Larget B, Liu L, Suchard MA, Huelsenbeck

JP (2012) MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model

space. Syst Biol 61:539-542

Ros AF, Lusa J, Meyer M, Soares MC, Oliveira RF, Brossard M, Bshary R (2011) Does access to the bluestreak cleaner

wrasse Labroides dimidatus affect indicators of stress and health in resident reef fishes in the Red Sea? Horm

Behav 59:151-158

Russell BC (1983) Annotated checklist of the coral reef fishes in the Capricorn-Bunker group Great Barrier Reef,

Australia. Great Barrier Reef Marine Park Authority, Mackay

Sambrook J, Russell DW (2001) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold

Spring, New York

Sasal P, Morand S, Guégan JF (1997) Determinants of parasite species richness in Meditteranean marine fishes. Mar

Ecol Prog Ser 149:61-71

Scholz T (2002) Family Gymnophallidae Odhner, 1905. In: Gibson DI, Jones A, Bray RA (eds) Keys to the Trematoda,

Volume 1. CABI Publishing and the Natural History Museum, Wallingford,

Scott ME, Dobson A (1989) The role of parasites in regulating host abundance. Parasitol Today 5:176-183

Sikkel PC, Cheney KL, Côté IM (2004) In situ evidence for ectoparasites as a proximate cause of cleaning interactions

in reef fish. Anim Behav 68:241-247

Sikkel PC, Schaumburg CS, Mathenia JK (2006) Diel infestation dynamics of gnathiid isopod larvae parasitic on

Caribbean reef fish. Coral Reefs 25:683-689

Sikkel PC, Ziemba RE, Sears WT, Wheeler JC (2009) Diel ontogenetic shift in parasitic activity in a gnathiid isopod on

Caribbean coral reefs. Coral Reefs 28:489-495

Sikkel PC, Sears WT, Weldon B, Tuttle BC (2011) An experimental field test of host-finding mechanisms in a

Caribbean gnathiid isopod. Mar Biol 158:1075-1083

Simpson SD, Jeffs A, Montgomery JC, McCauley R, Meekan MG (2008) Nocturnal relocation of adult and juvenile

coral reef fishes in response to reef noise. Coral Reefs 27:97-104

Smit NJ, Davies AJ (2004) The curious life-style of the parasitic stages of gnathiid isopods. Adv Parasitol 58:289-391

http://www.r-project.org)/

112

Smit NJ, Basson L, Van As JG (2003) Life cycle of the temporary fish parasite, Gnathia africana (Crustacea: Isopoda:

Gnathiidae). Folia Parasitol (Praha) 50:135-142

Smith JW (2002) Family Sanguinicolidae Graff, 1907. In: Gibson DI, Jones A, Bray RA (eds) Keys to the digenean

parasites of vertebrates. CAB International, Cambridge,

Soares MC, Bshary R, Côté IM (2008) Does cleanerfish service quality depend on client value or choice options? Anim

Behav 76:123-130

Soares MC, Oliveira R, Ros AFH, Grutter AS, Bshary R (2011) Tactile stimulation lowers stress in fish. Nat Commun

2:534

Srinivasan M (2003) Depth distibutions of coral reef fishes: the influence of microhabitat structure, settlement, and

post-settlement processes. Oecologia 137:76-84

Stachowicz JJ (2001) Mutualism, facilitation, and the structure of ecological communities. BioScience 51:235-246

Stamatakis A, Hoover P, Rougemont J (2008) A rapid bootstrap algorithm for the RAxML web servers. Syst Biol

57:758-771

Stummer LE, Weller JA, Johnson ML, Côté IM (2004) Size and stripes: how fish clients recognize cleaners. Anim

Behav 68:145-150

Sun D, Blomberg SP, Cribb TH, McCormick MI, Grutter AS (2012) The effects of parasites on the early life stages of a

damselfish. Coral Reefs 31:1065-1075

Sun D, Bray RA, Yong R, Q-Y, Cutmore SC, Cribb TH (2014) Pseudobacciger cheneyae n. sp. (Digenea:

Gymnophalloidea) from Weber’s chromis (Chromis weberi Fowler & Bean) (Perciformes: Pomacentridae) at

Lizard Island, Great Barrier Reef, Australia. Syst Parasitol 88:141-152

Sweatman H (1985) The influence of adults of some coral reef fishes on larval recruitment. Ecol Monogr 55:469-485

Sweatman H (1988) Field evidence that settling coral reef fish larvae detect resident fishes using dissolved chemical

cues. J Exp Mar Biol Ecol 124:163-174

Tolimieri N (1995) Effects of microhabitat characteristics on the settlement and recruitment of a coral reef fish at two

spatial scales. Oecologia 102:52-63

Toman G, Kamegai S (1974) Some digenetic trematodes of marine fishes from Saipan, Micronesia Proceedings of the

Third International Congress of Parasitology, Munich 1617-1618

Torres J, Miquel J, Casanova J-C, Ribas A, Feliu C, Morand S (2006) Endoparasite species richness of Iberian

carnivores: influences of host density and range distribution. Biodivers Conserv 15:4619-4632

Vail AL, McCormick MI (2011) Metamorphosing reef fishes avoid predator scent when choosing a home. Biol Lett

7:921-924

Waldie PA, Blomberg SP, Cheney KL, Goldizen AW, Grutter AS (2011) Long-term effects of the cleaner fish

Labroides dimidiatus on coral reef fish communities. PLoS ONE 6:e21201-21201-e21201-21207

Wang P, Wang Y (1993) Some digenetic trematodes from marine fishes from Fujian Province. Journal of Fujian

Teachers University Natural Science Edition 9:66-73

Weeks P (2000) Red-billed oxpeckers: vampires or tickbirds? Behav Ecol 11:154-160

Wenger AS, Johansen J, Jones G (2011) Suspended sediment impairs habitat choice and chemosensory discrimination

in two coral reef fishes. Coral Reefs 30:879-887

Whittington ID (1998) Diversity "down under": monogeneans in the Antipodes (Australia) with a prediction of

monogenean biodiversity worldwide. Int J Parasitol 28:1481-1493

Wilson SK, Burgess SC, Cheal AJ, Emslie M, Fisher R, Miller I, Polunin NVC, Sweatman HPA (2008) Habitat

utilization by coral reef fish: implications for specialists vs. generalists in a changing environment. J Anim

Ecol 77:220-228

Yamaguti S (1934) Studies on the helminth fauna of Japan. Part 2. Trematodes of fishes. Japanese Journal of Zoology

5:249-541

Yamaguti S (1938) Studies on the Helminth Fauna of Japan. Part 21. Trematodes of Fishes, IV. Yamaguti, S., Kyöto

Yamaguti S (1942) Studies on the helminth fauna of Japan. Part 39. Trematodes of fishes mainly from Naha.

Transactions of the Biogeographical Society of Japan 3:329-398

Yamaguti S (1953) Parasitic worms mainly from Celebes. Part 3. Digenetic trematodes of fishes II. Acta Med

Okayama 8:257-295

Yeo SE, Spieler RE (1980) Habitat effects on the occurence of parasites inhabiting the Sergeant Major, Abudefduf

saxatilis (Linnealus), with a list of parasites of Caribbean damselfish. Bull Mar Sci 30:313-324

113

Appendices

Appendix I

All damselfish recruits observed on experimental patch reefs (Chapter 3).

Fish Genus Species

Acanthochromis polyacanthus

Amblyglyphidodon curacao

Amblyglyphidodon leucogaster

Chromis ternatensis

Chromis viridis

Chrysiptera cyanea †

Chrysiptera flavipinnis

Chrysiptera rex *

Chrysiptera rollandi

Chrysiptera talboti

Dascyllus aruanus

Dascyllus reticulatus

Dischistodus perspicillatus

Dischistodus prospotaenia

Dischistodus spp.

Neoglyphidodon melas

Neoglyphidodon nigroris

Neopomacentrus bankieri

Neopomacentrus cyanomos

Pomacentrus adelus

Pomacentrus amboinensis

Pomacentrus bankanensis †

Pomacentrus branchialis

Pomacentrus chrysurus

Pomacentrus coelestis

Pomacentrus

Pomacentrus

grammorhynchus

lepidogenys

Pomacentrus moluccensis

Pomacentrus nagasakiensis

Pomacentrus pavo

Pomacentrus simsiang

* Species only recorded at reefs with L. dimidiatus present.

† Species only recorded at reefs without L. dimidiatus present.

114

Appendix II

Species of fish cleaned by juvenile Labroides dimidiatus (Chapter 4).

Species Family

Ctenochaetus binotatus

Acanthuridae

Ctenochaetus striatus Acanthuridae

Zebrasoma scopas Acanthuridae

Zebrasoma veliferum Acanthuridae

Apogon doederleini Apogonidae

Apogon compressus Apogonidae

Apogon cookii Apogonidae

Apogon properupta Apogonidae

Apogon trimaculatus Apogonidaae

Cheilodipterus intermedius Apogonidae

Cirripectes chelomatus Blennidae

Cirripectes stigmaticus Blennidae

Ecsenius aequalis Blennidae

Ecsenius bicolor Blennidae

Ecsenius mandibularis Blennidae

Ecsenius stictus Blennidae

Exallias brevis Blennidae

Meiacanthus atrodorsalis Blennidae

Plagiotremus rhinorhynchos Blennidae

Plagiotremus tapeinosoma Blennidae

Caesio caerulaurea Caesionidae

Caesio teres Caesionidae

Chaetodon aureofasciatus Chaetodontidae

Chaetodon baronessa Chaetodontidae

Chaetodon citrinellus Chaetodontidae

Chaetodon klienii Chaetodontidae

Chaetodon lunulatus Chaetodontidae

Chaetodon plebius Chaetodontidae

Chaetodon rainfordi Chaetodontidae

Chaetodon ulietensis Chaetodontidae

Amblygobius rainfordi Gobidae

Valenciennea strigata Gobidae

Neoniphon opercularis Holocentridae

Neoniphon sammara Holocentridae

Sarocentron cornutum Holocentridae

Sargocentron diadema Holocentridae

Sargocentron spiniferum Holocentridae

Bodianus axillaris Labridae

Cheilinus chlorourus Labridae

Chlorurus sordidus Labridae

Choerodon fasciatus Labridae

Coris batuensis Labridae

115

Epibulis insidiator Labridae

Gomphosus varius Labridae

Halichoeres hortulanus Labridae

Halichoeres malanurus Labridae

Halichoeres melanurus Labridae

Hemigymnus melapterus Labridae

Labrichtys unilineatus Labridae

Labroides dimidiatus Labridae

Pseudocheilinus hexataenia Labridae

Thalassoma lunare Labridae

Scarus frenatus Labridae

Scarus ghobban Labridae

Scarus niger Labridae

Stethojulis bandanensis Labridae

Lutjanus carponotatus Lutjanidae

Pervagor janthinosoma Monacanthidae

Parupeneus multifasciatus Mullidae

Scolopsis bilineatus Nemipteridae

Centropyge bicolor Pomacanthidae

Pomacanthus sexstriatus Pomacanthidae

Abudefduf sexfaciatus Pomacentridae

Abudefduf whitleyi Pomacentridae

Acanthochromis polyacanthus Pomacentridae

Amblyglyphididon curacao Pomacentridae

Amblyglyphididon leucogaster Pomacentridae

Chromis actripectoralis Pomacentridae

Chromis lepidolepis * Pomacentridae

Chromis viridis Pomacentridae

Chrysiptera rex Pomacentridae

Dischistodus melanotus Pomacentridae

Dischistodus prosopotaenia Pomacentridae

Dischistodus pseudochrysopoecilus Pomacentridae

Hemiglyphidodon plagiometopon Pomacentridae

Neoglyphidodon melas Pomacentridae

Neopomacentrus azysron Pomacentridae

Plectroglyphidodon lacrymatus Pomacentridae

Pomacentrus adelus * Pomacentridae

Pomacentrus amboinensis * Pomacentridae

Pomacentrus bankanensis Pomacentridae

Pomacentrus branchialis * Pomacentridae

Pomacentrus coelestis Pomacentridae

Pomacentrus chrysurus Pomacentridae

Pomacentrus grammorhynchus Pomacentridae

Pomacentrus imitator Pomacentridae

Pomacentrus moluccensis Pomacentridae

Pomacentrus nagasakiensis * Pomacentridae

Pomacentrus simsiang Pomacentridae

Pomacentrus wardi Pomacentridae

Premnas biaculeatus Pomacentridae

Stegastes apicalis Pomacentridae

116

Stegastes fasciolatus Pomacentridae

Stegastes nigricans Pomacentridae

Ogilbyina queenslandia Pseudochromidae

Pseudochromis fuscus Pseudochromidae

Cephalopholis boenak Epinephelidae

Cephalopholis cyanostigma Epinephelidae

Cephalopholis microprion Epinephelidae

Siganus corallinus Siganidae

Synodus variegatus Synodontidae

Canthigaster papua Tetraodontidae

Canthigaster valentini Tetraodontidae

* Recently-settled individuals (< 2 cm) of these species were cleaned by L. dimidiatus.

Documents

Parasites and cleaning behaviour in damselfishes Derek Sun