Moorea BIOCODE barcode library as a tool for understanding predator-prey interactions: Insights into the diet of common predatory coral reef fishes

Abstract

Identifying species involved in consumer–resource interactions is one of the main limitations in the construction of food webs. DNA barcoding of prey items in predator guts provides a valuable tool for characterizing trophic interactions, but the method relies on the availability of reference sequences to which prey sequences can be matched. In this study, we demonstrate that the COI sequence library of the Moorea BIOCODE project, an ecosystem-level barcode initiative, enables the identification of a large proportion of semi-digested fish, crustacean and mollusks found in the guts of three Hawkfish and two Squirrelfish species. While most prey remains lacked diagnostic morphological characters, 94% of the prey found in 67 fishes had >98% sequence similarity with BIOCODE reference sequences. Using this species-level prey identification, we demonstrate how DNA barcoding can provide insights into resource partitioning, predator feeding behaviors and the consequences of predation on ecosystem function.

Full text

1 23
Coral Reefs
Journal of the International Society for
Reef Studies
ISSN 0722-4028
Coral Reefs
DOI 10.1007/s00338-011-0845-0
Moorea BIOCODE barcode library as
a tool for understanding predator–prey
interactions: insights into the diet of
common predatory coral reef fishes
M.Leray, J.T.Boehm, S.C.Mills &
C.P.Meyer

1 23
Your article is protected by copyright and
all rights are held exclusively by Springer-
Verlag. This e-offprint is for personal use only
and shall not be self-archived in electronic
repositories. If you wish to self-archive your
work, please use the accepted author’s
version for posting to your own website or
your institution’s repository. You may further
deposit the accepted author’s version on a
funder’s repository at a funder’s request,
provided it is not made publicly available until
12 months after publication.

NOTE
Moorea BIOCODE barcode library as a tool for understanding
predator–prey interactions: insights into the diet of common
predatory coral reef fishes
M. Leray •J. T. Boehm •S. C. Mills •
C. P. Meyer
Received: 15 August 2011 / Accepted: 7 November 2011
ÓSpringer-Verlag 2011
Abstract Identifying species involved in consumer–
resource interactions is one of the main limitations in the
construction of food webs. DNA barcoding of prey items in
predator guts provides a valuable tool for characterizing
trophic interactions, but the method relies on the avail-
ability of reference sequences to which prey sequences
can be matched. In this study, we demonstrate that the
COI sequence library of the Moorea BIOCODE project,
an ecosystem-level barcode initiative, enables the identifi-
cation of a large proportion of semi-digested fish, crusta-
cean and mollusks found in the guts of three Hawkfish and
two Squirrelfish species. While most prey remains lacked
diagnostic morphological characters, 94% of the prey
found in 67 fishes had [98% sequence similarity with
BIOCODE reference sequences. Using this species-level
prey identification, we demonstrate how DNA barcoding
can provide insights into resource partitioning, predator
feeding behaviors and the consequences of predation on
ecosystem function.
Keywords Trophic interactions Diet analysis 
Food web DNA identification Hawkfish Squirrelfish
Introduction
The high biodiversity of coral reefs means that ecologists
are confronted with a complex task of species identification
in their quest for understanding community-level processes
and interactions. Subtle differences in diagnostic pheno-
typic characters, presence of morphologically cryptic and
undescribed species, and lack of identification guides for
early life stages hinder reliable species-level identification
in routine ecological studies (Hebert et al. 2003). Fortu-
nately, DNA barcoding can be used to supplement tradi-
tional taxonomy when their DNA matches species-specific
sequences available in barcode reference libraries.
Witnessing direct predator–prey interactions in the field
is challenging (Merfield et al. 2004); therefore, DNA-based
techniques are increasingly used for characterizing
predator diet from feces/gut content (King et al. 2008).
Prey-specific DNA fragments can be amplified from semi-
digested prey (Zaidi et al. 1999; Dunn et al. 2010), and
prey sequences can be identified if reference barcode dat-
abases contain a comprehensive list of species consumed.
Despite the growing availability of reference databases,
large proportions remain unidentified particularly from
generalist diets (Blankenship and Yayanos 2005; Dunn
et al. 2010).
Among various ongoing barcoding initiatives, the
Moorea BIOCODE project (http://www.mooreabiocode.org)
is an ‘‘All Taxa Biotic Inventory’’ whose goal is to provide
Communicated by Biology Editor Dr. Hugh Sweatman
M. Leray (&)S. C. Mills
Laboratoire d’Excellence ‘‘Corail’’ USR 3278 CNRS EPHE,
CRIOBE-CBETM, Universite
´de Perpignan,
56 Avenue Paul Alduy, 66860 Perpignan Cedex, France
e-mail: leray.upmc@gmail.com
M. Leray C. P. Meyer
Department of Invertebrate Zoology, National Museum
of Natural History, Smithsonian Institution, P.O. Box 37012,
MRC-163, Washington, DC 20013, USA
J. T. Boehm
Biology Department, Queens College,
City University of New York, Flushing, NY 11367, USA
J. T. Boehm
The Graduate Center, City University of New York,
New York, NY 10016, USA
123
Coral Reefs
DOI 10.1007/s00338-011-0845-0
Author's personal copy

a library of genetic markers for all non-microbial species of
the French Polynesian tropical ecosystem. From 2006 to
2010, teams of researchers have worked to sample mac-
robiotic species ([5,670 species C2 mm) of which 3,877
(68%) are coral reef species. All specimens were identified
morphologically to lowest taxon level, photographed and
their tissue sampled for DNA barcoding. A library of
species-specific DNA signatures amplified from a single
homologous region, the cytochrome coxidase subunit I,
was constructed for most animals. Reference specimens
were sent to museum collections. All information, from the
collection of specimens to their sequencing, was central-
ized in BIOCODE’s field and laboratory information
management systems. As of April 2011, reference data
exist for 28 marine phyla, with an emphasis on arthropods,
chordates and mollusks (http://biocode.berkeley.edu). The
barcode inventory of this model ecosystem will allow
researchers to overcome many limitations inherent in
morphology-based identification when species-level infor-
mation is required, for example to understand predator
feeding ecology and food web dynamics.
Here, we used direct sequencing to identify prey
remains in the stomachs of five common predator fish on
Moorean reefs with contrasting feeding regimes: three
hawkfish, Paracirrhites arcatus,P.forsteri and P.hemis-
tictus (Order: Perciformes; Family: Cirrhitidae), and two
squirrelfish, Sargocentron microstoma and S.tiere (Order:
Beryciformes; Family: Holocentridae). The three hawkfish
species commonly occupy coral colonies of the genus
Pocillopora where they sit and wait for prey during the day
(Kane et al. 2009). In contrast, the two squirrelfish species
actively look for benthic prey at night (Arias-Gonzalez
et al. 1998; Randall 2005). We aimed to: (1) assess the
proportion of prey that matched BIOCODE reference
sequences in order to evaluate the efficacy of BIOCODE’s
efforts to inventory macro-invertebrates and fishes and
(2) investigate how species-level prey identification could
provide insights into resource partitioning and feeding
behaviors in relation to the life history traits of predators as
well as into the consequences of predation on ecosystem
function. As this is the first attempt to characterize the diet
of coral reef-associated predators using DNA-based tech-
niques, this approach provides great promise for under-
standing complex trophic interactions.
Materials and methods
Fish collection and gut content dissection
A total of 67 adult carnivorous fish (33 P.arcatus,11
P.forsteri,7P.hemistictus,8S.microstoma and 8 S.tiere)
were speared on the north shore forereef of Moorea, French
Polynesia (17°300S, 149°500W), during the Austral Winters
of 2009 and 2010. The diurnal species (Paracirrrhites spp.)
were sampled at dusk, while the nocturnal predators
(Sargocentron spp.) were collected both at dawn and 2 hrs
after dusk. We did not observe prey regurgitation while
capturing predators. Fishes were preserved in cold 50%
ethanol in situ. In the laboratory, stomach contents were
dissected, and all visually distinguishable prey items
identified to the lowest taxon possible based on morphol-
ogy. Tissue samples were then taken from these prey
remains, rinsed with distilled water, counted and placed in
individual tubes for extraction and barcoding. Remaining
stomach contents were discarded, which is likely to have
biased our results toward larger and hard prey items that
better resist digestion.
DNA analysis and sequence identification
Total genomic DNA was extracted using automated
phenol–chloroform extraction with the Autogenprep 965
(Autogen, MA) with a final elution volume of 100 ll. COI
fragments were PCRed as 20-ll reactions with 0.6 llof
10 lM of each universal forward and reverse primers
(Folmer et al. 1994), 0.2 ll of Biolase taq polymerase
(Bioline) 5 U ll
-1
, 0.8 llof50mMMg
2?
,1llof10lM
dNTP and 1 ll of genomic DNA. PCR conditions were as
follows: 5 min at 95°C; 35 cycles of 30 s at 95°C; 30 s at
48°C; 45 s at 72°C; and a final 5 min at 72°C. Sequences
were identified based on similarity to the BIOCODE
sequence library using BLAST (Altschul et al. 1997)
searches performed in Geneious Pro 5.0.3 (Biomatters).
COI sequences were then assigned to taxonomic groups
according to criteria defined by Machida et al. (2009) and
Plaisance et al. (2009). Sequences were considered to
match reference specimens when sequence similarity was
[98%. In order to test whether our sampling effort was
sufficient, expected species accumulation curves with
95% confidence intervals were computed using EstimateS
(Colwell et al. 2004).
Results and discussion
Of the 67 fish speared, 52 had visually distinguishable prey
remains in their stomach encompassing 102 total individual
prey items. The majority of fish had only one visually
distinguishable prey item, but number of prey items ranged
from 0 to 7 (Fig. 1). Based on the size of observed hard
parts, all crustacean and mollusks were consumed as adults
while fish had been preyed upon as juveniles. Morpho-
logical identification of crustacean appendages (62 items)
and fish fins (23 items) rarely provided identifications
lower than the family level. Only two crustacean prey
Coral Reefs
123
Author's personal copy

items from two P.arcatus were identified to the species
level (Menaethius monoceros). Morphological identifica-
tion therefore achieved less than 2% success at species
level, later verified by DNA.
On the other hand, COI sequences obtained from 96
(of 102) prey items showed higher than 98% levels of
sequence similarity with BIOCODE reference sequences
(Table 1). In comparison, only 16 prey items (8 species)
had less than 98% similarity with sequences in GenBank
(excluding BIOCODE-generated sequences—Table 1).
Of the four remaining crustacean sequences without [98%
matches, one was identified to the family Parthenopidae
(85% similarity), while three remaining crustaceans could
not be confidently assigned to any taxonomic group (\80%
similarity). Two sequences matching bacterial DNA frag-
ments were discarded. Overall, 94% of the sequences were
identified to species level using COI sequences, demon-
strating the efficiency with which BIOCODE has sampled
fish, macro-crustaceans and macro-mollusks from the
Moorea reef community (GenBank accession numbers
JN107891–JN107990).
Despite the high level ([98%) of sequence similarity,
BIOCODE reference specimens could not provide species
names to 13 prey items: 1 fish, 10 crustacean and 2 mollusks.
These vouchered specimens either are undescribed species
(e.g., Galatheidae) or require further taxonomic work for
correct identification. Two prey items matched specimens
with identical names (Chlorodiella laevissima) but are
genetically distinct (K2P dist. ±SE =0.103 ±0.014),
suggesting the presence of cryptic lineages. Taxonomic
refinement and new species descriptions are ongoing as part
of the BIOCODE project. A large proportion of the crusta-
cean database is publically available on BOLD (Project
MBMIA), and fishes will be released upon final acceptance
of a manuscript under revision. All comparative data will be
made public by July 2012 according to funding obligations.
The rarefaction curve indicates that additional fish col-
lections would be required to better characterize the diet of
these predatory species (Fig. 2). Additionally, despite these
five species being known to consume mostly other fish,
crustacean and mollusks, our dietary analysis likely missed
soft-bodied prey species that can be detected using PCR
amplification and sequencing from the whole-gut content
tissue homogenate (Jarman et al. 2004; Deagle et al. 2009).
Despite the need for additional sampling of predator diet,
we discuss how prey identification to the species level
using barcoding can provide novel insights into resource
partitioning, predator feeding behaviors and the potential
consequences of predation on ecosystem function.
Firstly, the degree to which specialization on food
resources enables species coexistence and shapes commu-
nity structure has long been debated (Jones 1991). This
debate has been limited as estimates of the degree of diet
overlap between predator species greatly depends on
taxonomic resolution achieved in dietary studies (Longe-
necker 2007). Paracirrhites forsteri and P.hemistictus,
both large diurnal ambush piscivorous species commonly
found among the branches of large pocilloporid corals
(Kane et al. 2009), had narrow and largely overlapping
diets with a majority of Chromis vanderbilti (80 and 67%
of prey items, respectively) in their guts (Table 1; Fig. 3).
Alternatively, S.microstoma and S. tiere, both mobile
active nocturnal feeders (Arias-Gonzalez et al. 2004),
have broad diets and did not have a single prey species in
common (Fig. 3). Their different diets reinforce the value
of species-level prey identifications, as familial level would
have failed to elucidate the true trophic structure of these
sister species. Paracirrhites arcatus only shared one single
prey species with its congenerics, whereas eight prey spe-
cies were shared with predators that differ in microhabitat
use and time of feeding activity (Fig. 3). These results,
which must be treated with caution due to the limited
number of samples and potentially different digestion rates,
suggest that timing of activity, habitat partitioning and
hunting mode may not accurately predict resource parti-
tioning among reef fish species. The barcode inventory of
the Moorea ecosystem provides an ideal testing ground for
further exploration of the role that resource specialization
plays in shaping patterns of biodiversity.
Secondly, our findings indicate that a few prey species
may provide a considerable source of energy to predators
in the Moorea food web. For instance, C.vanderbilti was
commonly consumed by P. forsteri and P. hemistictus and
Liocarpilodes integerrimus by P. arcatus and S. micros-
toma (Table 1), suggesting that they may be preferential
targets or highly abundant on Moorean reefs.Empirical
evidence suggests that generalist piscivorous predators
forage non-selectively and consume prey in proportion to
their abundance (Heinlein et al. 2010—study in Moorea).
Fig. 1 Number of prey items from the guts of 67 fishes belonging to
five predator species. TL total length
Coral Reefs
123
Author's personal copy

Table 1 Summary of prey items successfully identified from the stomach contents of fish using COI amplification and BLAST searches in the
BIOCODE barcode library
Subphylum/class Prey ID % identity BIOCODE nBIOCODE specimen Predator ID
Actinopterygii Chromis acares 99 1 MParis0005 ST
Chromis iomelas 99.8 2 MParis0055 PH
Chromis vanderbilti 99–99.5 12 MParis0195 PA, PF, PH
Cirripectes variolosus 100 1 MParis0213 PA
Eviota disrupta 99.7 1 MParis0174 PA
Eviota sp. 99.4 1 XMOO-0047 PA
Neocirrhites armatus
a
99.2 2 MParis0007 PF, PH
Pseudogramma polyacanthum 99.5 1 MParis0012 ST
Synodus binotatus
a
98.8 1 MParis586 SM
Valenciennea strigata
a
100 1 MParis0906 SM
Crustacea Acanthanas pusillus 99.1 1 BMOO-02811 ST
Alpheus dolerus 99.7–100 2 BMOO-00430 PA
Aniculus retipes
a
100 1 BMOO-01743 ST
Axiidae 98.2 1 BMOO-01079 PA
Brachycarpus biunguiculatus 98 1 BMOO-05348 PA
Calappa gallus 100 1 BMOO-02116 PA
Chlorodiella barbata 100 2 BMOO-00324 PA, SM
Chlorodiella crispipleopa 99.5–99.7 3 BMOO-00726 PA
Chlorodiella laevissima 98.4 1 BMOO-01191 SM
Chlorodiella laevissima 99.8 3 BMOO-02899 ST
Cyclodius ungulatus 98–99.8 2 BMOO-01049 PA, SM
Daldorfia sp. 99.2 1 BMOO-05257 PA
Epialtidae 99.2 1 BMOO-03230 PA
Etisus frontalis 99.7 1 BMOO-00194 PA
Galathea mauritiana
a
99.5–99.8 2 XMOO-0011 PA, SM
Galathea sp.
a
99.5–99.8 4 BMOO-02353 PA, ST
Gnathiidae 98.5 1 No voucher PA
Gonodactylus affinis 99.4 1 jg8 PA
Huenia sp. 99.2 1 BMOO-03531 PA
Liocarpilodes integerrimus 98.8–99.8 7 BMOO-01576 PA, SM
Medaeus elegans 99.4 1 BMOO-04008 PA
Menaethius monoceros 98.5–99.5 2 BMOO-03072 PA
Menaethius orientalis 99 1 BMOO-01847 ST
Metalpheus nanus 99.7 1 BMOO-02919 PA
Palaemonella rotumana 100 1 BMOO-02250 PA
Palmyria palmyrensis 100 1 BMOO-01358 PA
Parthenopidae 85 1 No match ST
Perinia tumida 98.5 1 XMOO-0383 PA
Petrolisthes sp. 98.7% 1 BMOO-02308 PA
Phylladiorhynchus sp. 99.6 1 BMOO-03262 PA
Phylladiorhynchus integrirostris 99.2–99.7 4 BMOO-01856 PA
Pilodius flavus 99.5–100 4 BMOO-02998 PA, ST
Pilodius pugil 100 2 BMOO-01102 SM
Saron marmoratus 99 1 BMOO-02912 ST
Saron sp. 98.3 1 No voucher PA
Thalamita sp. 98 2 BMOO-05357 PA, SM
Trapezia flavopunctata 99.7 1 BMOO-02830 ST
Trapezia tigrina 100 2 XMOO-0202 PA
Coral Reefs
123
Author's personal copy

Conversely, Longenecker (2007) observed a different pat-
tern for predators; despite large ephemeral increases in the
abundance of certain invertebrate prey species, they were
not increasingly consumed by predators. Invertebrate
population sizes, as well as temporal and spatial patterns
of variation in abundance, remain unknown in Moorea.
Therefore, further studies should be conducted to evaluate
diet selection and the role keystone prey species (Power
et al. 1996) play in the persistence of coral reef predators.
Finally, DNA barcoding revealed that the fish predators
feed on prey which themselves are important for habitat
maintenance and ecosystem functioning. Paracirrhites
forsteri and P.hemistictus consumed Neocirrhites armatus,
and P.arcatus had fed upon Trapezia flavopunctata and
T.tigrina, which are all known to benefit Pocilloporids.
Resident fish such as N.armatus provide nutrients to host
polyps (Holbrook et al. 2008), while Trapezia species
increase the survival and growth of their host by removing
sediment from coral tissue (Stewart et al. 2006), defending
against corallivorous seastars (Glynn 1983) and removing
parasitic vermetid gastropod nets (Stier et al. 2010).
Further investigation should determine the functional
consequences resulting from the predation pressure high-
lighted in this study.
Overall, we show that the quality of the barcode refer-
ence database in Moorea will enable researchers to uncover
the complexity and spatial–temporal dynamics of food
webs not just in French Polynesia but also throughout the
Western Pacific where taxon ranges likely extend. DNA
barcoding removes subjectivity biasing prey identification
compared to visual identification and is particularly valu-
able for reef fish prey identification given the high eco-
system biodiversity. Such promises for understanding the
functioning of natural systems should encourage further
ecosystem-based barcoding initiatives worldwide.
Fig. 2 Rarefaction curve for number of prey species as a function of
fish collected with prey in their stomach (N=52). Dashed lines
represent 95% CI
Fig. 3 Venn diagram illustrating the overlap of prey species
consumed by predator fish species. Circle sizes are proportional to
total number of prey species consumed by each predator (parenthe-
ses), and overlap area between circles is proportional to number of
shared prey species. Photographs were provided by Jeffrey Williams
Table 1 continued
Subphylum/class Prey ID % identity BIOCODE nBIOCODE specimen Predator ID
Xanthias latifrons 99.5 1 BMOO-04066 ST
Mollusk Deniatys dentifer
a
98 3 BMOO-02659 SM
Erato sp.
a
99.2 2 BMOO-02545 ST
Julia zebra 98.1 1 BMOO-03193 PA
Stomatella rosaceus 99.4 1 BMOO-01483 ST
Stomatolina sp. 99.5 1 BMOO-06132 ST
The BIOCODE reference specimen ID is reported for each prey so that photographs and additional information can be obtained at
http://biocode.berkeley.edu
n=number of prey. PA: Paracirrhites arcatus; PF: Paracirrhites forsteri; PH: Paracirrhites hemisticus; ST: Sargocentron tiere; SM:
Sargocentron microstoma
a
Indicates prey COI sequences that also had [98% similarity with sequences in GenBank
Coral Reefs
123
Author's personal copy

Acknowledgments We thank the BIOCODE teams who collected
marine invertebrates and fish specimen in 2006, the ‘‘Centre de
Recherche Insulaire et Observatoire de l’Environnement (CRIOBE)
de Moorea’’, the Richard B. Gump field station in Moorea for
logistical support and three anonymous reviewers for helpful com-
ments on the manuscript. We also greatly acknowledge the Gordon
and Betty Moore Foundation, Smithsonian Institution Fellowship
Program and France American Cultural Exchange program (FACE—
Partner University Fund) for financial support.
References
Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller
W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new
generation of protein database search programs. Nucleic Acids
Res 25:3389–3402
Arias-Gonzalez JE, Hertel O, Galzin R (1998) Fonctionnement
trophique d’un e
´cosyste
`me re
´cifal en Polyne
´sie franc¸aise.
Cybium 22:1–24
Arias-Gonzalez JE, Galzin R, Harmelin-Vivien M (2004) Spatial,
ontogenetic, and temporal variation in the feeding habits of the
squirrelfish Sargocentron microstoma on reefs in Moorea,
French Polynesia. Bull Mar Sci 75:473–480
Blankenship LE, Yayanos AA (2005) Universal primers and PCR of
gut contents to study marine invertebrate diets. Mol Ecol 14:
891–899
Colwell RK, Mao CX, Chang J (2004) Interpolating, extrapolating,
and comparing incidence-based species accumulation curves.
Ecology 85:2717–2727
Deagle BE, Kirkwood R, Jarman SN (2009) Analysis of Australian
fur seal diet by pyrosequencing prey DNA in faeces. Mol Ecol
18:2022–2038
Dunn MR, Szabo A, McVeagh MS, Smith PJ (2010) The diet of
deepwater sharks and the benefits of using DNA identification of
prey. Deep-Sea Res Part I 57:923–930
Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA
primers for amplification of mitochondrial cytochrome C oxidase
subunit I from diverse metazoan invertebrates. Mol Mar Biol
Biotechnol 3:294–299
Glynn PW (1983) Increased survivorship in corals harboring crusta-
cean symbionts. Mar Biol Lett 4:105–111
Hebert PDN, Cywinska A, Ball SL, DeWaard JR (2003) Biological
identifications through DNA barcodes. Proc R Soc London Ser
B-Biol Sci 270:313–321
Heinlein JM, 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
Holbrook SJ, Brooks AJ, Schmitt RJ, Stewart HL (2008) Effects of
sheltering fish on growth of their host corals. Mar Biol 155:
521–530
Jarman SN, Deagle BE, Gales NJ (2004) Group-specific polymerase
chain reaction for DNA-based analysis of species diversity and
identity in dietary samples. Mol Ecol 13:1313–1322
Jones GP (1991) Postrecruitment processes in the ecology of coral
reef fish populations: a multifactorial perspective. In: Sale PF
(ed) The ecology of fishes on coral reefs. Academic Press,
New York, pp 294–328
Kane C, Brooks A, Holbrook S, Schmitt R (2009) The role of
microhabitat preference and social organization in determining
the spatial distribution of a coral reef fish. Environ Biol Fish
84:1–10
King RA, Read DS, Traugott M, Symondson WOC (2008) Molecular
analysis of predation: a review of best practice for DNA-based
approaches. Mol Ecol 17:947–963
Longenecker K (2007) Devil in the details: high-resolution dietary
analysis contradicts a basic assumption of reef-fish diversity
models. Copeia 3:543–555
Machida RJ, Hashiguchi Y, Nishida M, Nishida S (2009) Zooplank-
ton diversity analysis through single-gene sequencing of a
community sample. BMC Genomics 10:438. doi:10.1186/1471-
2164-10-438
Merfield CN, Wratten SD, Navntoft S (2004) Video analysis of
predation by polyphagous invertebrate predators in the labora-
tory and field. Biol Control 29:5–13
Plaisance L, Knowlton N, Paulay G, Meyer C (2009) Reef-associated
crustacean fauna: biodiversity estimates using semi-quantitative
sampling and DNA barcoding. Coral Reefs 28:977–986
Power ME, Tilman D, Estes JA, Menge BA, Bond WJ, Mills LS,
Daily G, Castilla JC, Lubchenco J, Paine RT (1996) Challenges
in the quest for keystones. Bioscience 46:609–620
Randall J (2005) Reef and shore fishes of the South Pacific.
University of Hawaii Press, Honolulu
Stewart HL, Holbrook SJ, Schmitt RJ, Brooks AJ (2006) Symbiotic
crabs maintain coral health by clearing sediments. Coral Reefs
25:609–615
Stier AC, McKeon CS, Osenberg CW, Shima JS (2010) Guard crabs
alleviate deleterious effects of vermetid snails on a branching
coral. Coral Reefs 29:1019–1022
Zaidi RH, Jaal Z, Hawkes NJ, Hemingway J, Symondson WOC
(1999) Can multiple-copy sequences of prey DNA be detected
amongst the gut contents of invertebrate predators? Mol Ecol
8:2081–2087
Coral Reefs
123
Author's personal copy