Evidence from the Polypipapiliotrematinae N

Home , Allopodocotyle, Chaetodon rainfordi, Chaetodon trichrous, Fourspot butterflyfish, Hamacreadium, Heniochus monoceros

Accepted Manuscript

Intermediate host switches drive diversification among the largest trematode family: evidence from the Polypipapiliotrematinae n. subf. (Opecoelidae), parasites transmitted to butterflyfishes via predation of coral polyps

Storm B. Martin, Pierre Sasal, Scott C. Cutmore, Selina Ward, Greta S. Aeby, Thomas H. Cribb

PII: S0020-7519(18)30242-X DOI: https://doi.org/10.1016/j.ijpara.2018.09.003 Reference: PARA 4108

To appear in: International Journal for Parasitology

Received Date: 14 May 2018 Revised Date: 5 September 2018 Accepted Date: 6 September 2018

Please cite this article as: Martin, S.B., Sasal, P., Cutmore, S.C., Ward, S., Aeby, G.S., Cribb, T.H., Intermediate host switches drive diversification among the largest trematode family: evidence from the Polypipapiliotrematinae n. subf. (Opecoelidae), parasites transmitted to butterflyfishes via predation of coral polyps, International Journal for Parasitology (2018), doi: https://doi.org/10.1016/j.ijpara.2018.09.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Intermediate host switches drive diversification among the largest trematode family: evidence from the Polypipapiliotrematinae n. subf. (Opecoelidae), parasites transmitted to butterflyfishes via predation of coral polyps

Storm B. Martina,*, Pierre Sasalb,c, Scott C. Cutmorea, Selina Warda, Greta S. Aebyd, Thomas H.

Cribba

aSchool of Biological Sciences, The University of Queensland, Brisbane, QLD, 4072, Australia bPSL Research University, USR 3278 EPHE-UPVD-CNRS, Centre de Recherche Insulaire et

Observatoire de l’Environnement (CRIOBE) Université de Perpignan Via Domitia, 58 avenue P.

Alduy, 66860 Perpignan. France cLaboratoire d’excellence Corail, EPHE, Moorea, French Polynesia dHawai‘i Institute of Marine Biology, Kāne‘ohe, Hawai‘i, United States of America

*Corresponding author. E-mail address: [email protected]

Abstract

Podocotyloides stenometra Pritchard, 1966 (Digenea: Opecoelidae) is the only trematode known to infect anthozoan corals. It causes disease in coral polyps of the genus Porites Link (Scleractinia:

Poritidae) and its life-cycle depends on ingestion of these polyps by butterflyfishes (Perciformes:

Chaetodontidae). This species has been reported throughout the Indo-Pacific, from the Seychelles to the Galápagos, but no study has investigated whether multiple species are involved. Here, we recollect P. stenometra from its type-host and type-locality, in Hawaiian waters, and describe four new species from examination of 768 butterflyfishes from French Polynesia. On the basis of morphology, phylogeny and life-history, we propose Polypipapiliotrema Martin, Cutmore & Cribb n. gen. and the Polypipapiliotrematinae Martin, Cutmore & Cribb n. subf., for P. stenometra

(Pritchard) n. comb., P. citerovarium Martin, Cutmore & Cribb n. sp., P. hadrometra Martin,

Cutmore & Cribb n. sp., P. heniochi Martin, Cutmore & Cribb n. sp., and P. ovatheculum Martin,

Cutmore & Cribb n. sp. Given the diversity uncovered here and the ubiquity, abundance and diversity of butterflyfishes on coral reefs, we predict that Polypipapiliotrema will prove to comprise a rich complex of species causing disease in corals across the Indo-Pacific. The unique life-cycle of these taxa is consistent with phylogenetic distinction of the group and provides evidence for a broader basis of diversification among the family. We argue that life-cycle specialisation, in terms of adoption of disparate second intermediate host groups, has been a key driver of the diversification and richness of the Opecoelidae, the largest of all trematode families and the group most frequently encountered in coral reef fishes.

Keywords: Chaetodontidae, Corallivory, Host specificity, New species, Opecoelidae, Trematodiasis,

Taxonomy, Phylogeny

1. Introduction

Among the Trematoda, Podocotyloides stenometra Pritchard, 1966 (Opecoelidae Ozaki,

1925) is noteworthy as the only species demonstrated to cause disease in anthozoan corals (Aeby,

1998). Opecoelids have complex life-cycles involving three hosts and reach the definitive host, fishes, through trophic transmission. Other opecoelids are known to exploit bivalves, crustaceans, echinoids, gastropods, insects, annelids, or small fishes as second intermediate hosts (Cribb, 2005).

Cercariae of P. stenometra penetrate and encyst within coral polyp tissues, apparently only in species of Porites Link (Scleractinia: Poritidae), causing abnormal pink pigmentation and swelling such that the polyp may be unable to retract into its calyx (Cheng and Wong, 1974; Aeby, 1998).

This pathology renders polyps both more conspicuous and vulnerable to predators and thus corallivorous butterflyfishes (Chaetodontidae), the definitive hosts of P. stenometra, preferentially eat infected polyps (Aeby, 2002), perpetuating the life-cycle.

Considering its potential significance as a pathogen of reef-building corals, P. stenometra is surprisingly poorly known. It has been reported from butterflyfishes in Hawaiian (Pritchard, 1966),

French Polynesian (Martin et al., 2018c), Seychelloise (Toman, 1992) and Great Barrier Reef (Bray and Cribb, 1989; Lucas et al., 2005) waters. Additionally, cases of suspected Porites trematodiasis have been reported from Hawai‘i (Aeby, 2006, 2007), French Polynesia (M. Rigby in Aeby, 2007), the Great Barrier Reef (Willis et al., 2001), Guam and Papua New Guinea (Aeby, 2007), the Ryuku

Islands (Yamashiro, 2004), the Galapágos Archipelago (Vera and Banks, 2009) and the New

Caledonian lagoon (Aeby et al., 2015). Presently, these reports are all attributed to a single species.

Work et al. (2014) and Aeby et al. (2015) did note differences between infections in Hawaiian versus New Caledonian corals, but the possibility that multiple species are involved remains unexplored.

The taxonomic position of P. stenometra itself also requires re-evaluation. Pritchard (1966) placed P. stenometra in Podocotyloides Yamaguti, 1934 based on general morphological similarity to the type-species, P. petalophallus Yamaguti, 1934; both are elongate with a pedunculate ventral

3 sucker, an entire ovary and vitelline follicles restricted to the hindbody. However, its inclusion in the genus is unsatisfactory. The type species and other convincing congeners are known mainly from haemulids (Perciformes), generalist benthic carnivores which do not feed on coral polyps. In contrast to those species, P. stenometra lacks a uterine sphincter, a petalloid cirrus and a canalicular seminal receptacle, the latter a defining characteristic of not just Podocotyloides but also its nominal subfamily, the Plagioporinae Manter, 1947. Recent analyses of rDNA sequence data demonstrated that P. stenometra is distantly related to species of Podocotyloides (sensu stricto) and indeed is phylogenetically distinctive among all sequenced representatives of the Opecoelidae (Martin et al.,

2018c). Those analyses suggested P. stenometra requires a new genus, but were based on sequence data generated from infections in Heniochus chrysostomus Cuvier (Perciformes: Chaetodonidae) collected off Moorea, French Polynesia (Martin et al., 2018c), whereas the type combination for P. stenometra is Chaetodon quadrimaculatus Gray (Chaetodontidae) from Hawaiian waters (Pritchard,

1966).

In this study we recollect and provide sequence data for P. stenometra from its type combination and prospect for unrecognised diversity among butterflyfishes from French Polynesian waters. Specifically, we assess the evidence for the occurrence of P. stenometra (sensu stricto) outside of Hawaiian waters and discuss the significance of the adoption of coral hosts apparent in the life-cycle of this trematode from an evolutionary context.

2. Materials and methods

2.1. Specimen collection

Butterflyfishes were collected from Hawaiian and French Polynesian waters. In Hawai‘i, fishes were purchased from a professional collector operating from off Hale‘iwa, Waialua, O‘ahu in

June, 2016. Fishes from French Polynesian waters were collected mostly by microspear on snorkel or rotenone, across several expeditions to multiple localities (Fig. 1): from off Moorea, Society

Islands in November-December, 1999, November, 2009, November, 2012 and April, 2017; from off the Mangareva island group in the Gambier Islands in October, 2010; from off Eiao, Fatu Hiva,

Fatu Huku, Hiva Oa, Motu One, Nuku Hiva, Tahuata and Ua Pao, Marquesas Islands in October-

November, 2011; from off Maria, Raivavae, Rimatara, Rurutu and Tubuai, Austral Islands in April,

2013; and from off Fakarava and Toau in November-December, 2012 and off Rangiroa in April,

2017, Tuamotu Archipelago. Fishes were dissected fresh and intestinal parasites were collected as per Cribb and Bray (2010). Trematodes were fixed, without pressure, in near-boiling saline and preserved in 80% ethanol.

2.2. Morphological study

Specimens were stained and mounted as per Martin et al. (2017). Measurement data were taken from lateral mounts, using the software package cellSens Standard v1.13 via live feed from an

Olympus SC50 camera mounted onto an Olympus BX53 compound microscope. Measurements are in micrometres, expressed as length by depth (eggs as length by width), with ranges followed by the mean in parentheses. Morphometric data were explored and comparative plots were generated using

R v3.2.2 (R Core Team, 2015. R: A language and environment for statistical computing. R

Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.). Illustrations were produced with a drawing tube attachment. Illustrations and figures exported from R were digitised in Adobe Illustrator CS6. Specimens collected from fishes in Hawaiian waters were lodged in the Smithsonian National Museum of Natural History, USA, Invertebrate Zoology collection

(USNM) and those from fishes in French Polynesian waters in the Museum National d'Histoires

Naturelles, France (MNHN HEL).

2.3. Molecular and phylogenetic study

Genetic sequence data were generated for three ribosomal DNA (rDNA) markers: the second internal transcribed spacer region (ITS2 rDNA) and the small (18S) and large (28S)

5 ribosomal subunit RNA coding regions. These markers were targeted for compatibility with existing opecoelid sequence data available in GenBank, especially those identified as relating to P. stenometra (Lucas et al., 2005; Martin et al., 2018c). Genomic DNA extraction, amplification and sequencing protocols, including primers, are detailed in Martin et al. (2017). Generated ITS2 sequences were annotated using the ITS2 database (Keller et al., 2009).

The phylogenetic affinities of species represented by the examined material were determined through maximum liklihood and Bayesian inference analyses. A concatenated 28S +

18S alignment was constructed, comprising all such data for opecoelid taxa available in GenBank

(Table 1). Based on previous analyses (Littlewood et al., 2015; Bray et al., 2016; Martin et al.,

2018c), the following taxa were selected to comprise the outgroup: Gorgocephalus yaaji Bray &

Cribb, 2005 (Lepocreadioidea Odhner, 1905: Gorgocephalidae Manter, 1966), Preptetos caballeroi

Pritchard, 1960 (Lepocreadioidea: Lepocreadiidae Odhner, 1905), Stephanostomum pristis

(Deslongchamps, 1824) (Brachycladioidea Odhner, 1905: Acanthocolpidae Lühe, 1906) and

Zalophotrema hepaticum Stunkard & Alvey, 1929 (Brachycladioidea: Brachycladiidae Odhner,

1905). The 28S and 18S data were aligned separately using MUSCLE v3.7 (Edgar, 2004) implemented in MEGA v6 (Tamura et al., 2013) with ClustalW sequence weighting and UPGMB clustering for iterations 1 and 2. Alignments were trimmed to the shortest 50% of sequences and indels greater than a single base-position and affecting >5% of sequences were removed, amounting to about 3% and 2% of the 28S and 18S alignments, respectively. Thus, the final concatenated alignment considered 1,319 and 1,814 base positions of the 28S and 18S marker regions. Analyses were run through implementations of RAxML v8 (Stamatakis, 2014) and MrBayes v3.2.6 (Ronquist et al., 2012) in the CIPRES portal (Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the

CIPRES Science Gateway for inference of large phylogenetic trees. Proceedings of the Gateway

Computing Environments Workshop (GCE), 14 Nov. 2010. New Orleans, LA, USA pp 1–8.), for maximum likelihood and Bayesian inference, respectively, using the closest approximation of the

GTR + I + Γ model of evolution, as suggested by estimates from the Akaike information criterion

6 implemented in PartitionFinder v1.1.1 (Lanfear et al., 2012). The maximum likelihood analysis was run with 1,000 bootstrap pseudoreplicates and the Bayesian inference analysis with four chains sampled every 1,000 of 10,000,000 iterations and the first 2,500 samples discarded as burn-in, at which point the average standard deviation of split frequencies was < 0.005.

2.4. Data accessibility

Raw morphometric data underpinning species descriptions and comparisons, and the concatenated 28S + 18S rDNA alignment used in the phylogenetic analyses, are freely accessible at https://data.mendeley.com/datasets/tzr348ybx5/1.

3. Results

3.1. Recognition of species

A total of 797 individual chaetodontids were examined for intestinal helminths, comprising

29 from off O‘ahu and 768 from French Polynesian waters (Table 2). Infections of species broadly consistent with P. stenometra (sensu lato) were recovered from two of four species of Chaetodon collected off O‘ahu and from eight of 23 species of Chaetodon and two of two species of Heniochus collected in French Polynesian waters. No specimens were encountered in species of

Hemitaurichthys or Forcipiger. Overall, prevalence was low among fishes collected from French

Polynesian waters, especially so at Moorea, the most heavily sampled site. Infections were most readily detected in the Austral Islands and none were encountered in any fishes from the Marquesas

Archipelago.

Seven genotypes were uncovered from 33 generated ITS2 sequences. Only one genotype occurred in fishes collected off O‘ahu, and it was not detected in French Polynesian waters. It is interpreted here to represent P. stenometra (sensu stricto) because there is no indication that more than a single species is present among chaetodontids in Hawaiian waters and because the material examined includes specimens from the type host, C. quadrimaculatus. This genotype does not

7 match previously published sequence data available in GenBank and identified as relating to P. stenometra, DQ083434 (Lucas et al., 2005) and MF925404 (Martin et al., 2018c), but those data were generated from specimens recovered from Heniochus chrysostomus Cuvier collected on the

Great Barrier Reef and off Moorea, respectively. We interpret the six genotypes occurring in French

Polynesia as relating to four new species congeneric with P. stenometra. We describe these species below and propose a new genus and new subfamily, with P. stenometra as the type species, to accommodate them.

3.2. Taxonomy

3.2.1. Polypipapiliotrematinae Martin, Cutmore & Cribb n. subf. (Digenea: Plagiorchiida:

Opecoelidae Ozaki, 1925)

Diagnosis. Body narrow, elongate, subcylindrical. Forebody occupies about 10–20% of body length. Tegument smooth, thin. Oral sucker terminal, unspecialised. Ventral sucker pedunculate, larger than oral sucker. Prepharynx short, usually prominent. Pharynx unspecialised, may be larger or smaller than oral sucker. Oesophagus simple. Intestine bifurcates in forebody or dorsal to ventral sucker. Caeca blind, terminate well beyond testes near posterior extremity. Testes two, entire, smooth, tandem to slightly diagonal, contiguous or separate. Cirrus-sac present, long, thin, extending into hindbody posteriorly. Seminal vesicle internal, unipartite. Pars-prostatica present. Ejaculatory duct long, simple. Genital pore medial or almost so, prebifurcal. Ovary entire, smooth, medial or sinistro-submedial, situated at least 50% of body length from anterior extremity.

Seminal receptacle uterine. Laurer’s canal absent. Vitellarium follicular, restricted to hindbody.

Mehlis’ gland present. Uterus entirely anterior to ovary. Metraterm discernible, with or without terminal sphincter, with or without thickened chamber. Eggs operculate, tanned, unembryonated in utero, lacking polar protuberances or filaments. Excretory vesicle tubular, restricted to hindbody, extends well-anterior to ovary. Adults parasitic in intestine of corallivorous fishes, known only from species of Chaetodon and Heniochus (Chaetodontidae) and Zanclus cornutus (Zanclidae). Known

8 metacercariae endo-parasitic in anthozoans, known only from species of Porites Link (Scleractinia:

Poritidae).

3.2.1.1. Taxonomic summary

Type-genus: Polypipapiliotrema n. gen.

Other genera: none.

3.2.2. Polypipapiliotrema Martin, Cutmore & Cribb gen. nov.

Diagnosis. With characters of the subfamily.

3.2.2.1. Taxonomic summary.

Type-species: Polypipapiliotrema stenometra (Pritchard, 1966) n. comb.

Other species: P. heniochi n. sp.; P. hadrometra n. sp.; P. ovatheculum n. sp.; P. citerovarium n. sp.

Zoobank registration: http://zoobank.org/urn:lsid:zoobank.org.act:371C9799-A1B1-4EEB-

B6AA-80BCB91B2604

Etymology: Polypipapiliotrema is composed from the Latin words polypum, a coral or polyp, papilio, a butterfly, and trema, from trematode, because these trematodes infect corals and butterflyfishes. The genus is treated as neuter.

3.2.2.2. Remarks

Proposal of Polypipapiliotrema is required to accommodate P. stenometra and the new species detected here, because, as per the new analyses and previous investigation (Martin et al.,

2018c), these species are phylogenetically distant from the type species of Podocotyloides and its genuine congeners. The revised concept of Podocotyloides is restricted to species known mainly from haemulid fishes and characterised by a canalicular seminal receptacle, a petalloid-cirrus, a

9 sinistral genital pore and a uterine sphincter immediately prior to the genital atrium (Martin et al.,

2018c). The species considered here to belong in Polypipapiliotrema occur in corallivorous fishes, namely chaetodontids and Zanclus cornutus, possess a uterine seminal receptacle, a medial genital pore and do not have a petalloid cirrus. Some of these species do exhibit a distal thickening of the metraterm into a sphincter-like structure, but this does not appear to be homologous with that observed in species of Podocotyloides (sensu stricto). There is no existing available genus to accommodate these species, and no indication that any other opecoelids incorporate exploitation of anthozoans in their life-cycle.

3.2.3. Polypipapiliotrema stenometra (Pritchard, 1966) Martin, Cutmore & Cribb n. comb. (Fig. 2

A, B)

Synonyms: Podocotyloides stenometra Pritchard, 1966

Description [based on 8 lateral whole-mounts (Table 3)]. Body narrow, elongate, subcylindrical. Forebody short. Tegument smooth, about 3 thick. Oral sucker terminal, unspecialised, ellipsoidal with depth greater than length. Ventral sucker supported by broad, short peduncle, ellipsoidal with length usually greater than depth, larger than oral sucker. Prepharynx short, prominent. Pharynx unspecialised, subspherical with length usually slightly greater than depth, slightly larger than oral sucker. Oesophagus gently sinuous, surrounded by sparse gland cells. Intestine bifurcates dorsal or immediately anterior to ventral sucker. Caeca blind, terminate well beyond testes, usually equal.

Testes smooth, ellipsoidal with length greater than depth, similar in size with posterior testis usually slightly larger than anterior testis, tandem, ventral, separated by short gap of 4–49 (23) or just touching. Cirrus-sac thin, elongate, tapering slightly distally, gently curved to sigmoid, extending beyond ventral sucker into hindbody. Internal seminal vesicle narrow, unipartite. Pars prostatica narrow, elongate, distinct from seminal vesicle. Ejaculatory duct long, thin, muscular, simple. Genital atrium simple. Genital pore small, simple, ventral, medial or almost so, prebifurcal.

Ovary smooth, subspherical, medial, ventral, anterior to and separated by short gap [4–65

(23)] from anterior testis. Vitelline reservoir smaller than and antero-dorsal to ovary. Vitellarium composed of small, scattered follicles restricted to hindbody, distributed laterally and dorsally from posterior extremity to about level of seminal vesicle and ventrally in post-testicular region, may be irregularly interrupted. Oviduct short. Oötype well-defined. Mehlis’ gland diffuse, inconspicuous.

Uterus thin, entirely anterior to ovary, ventral and tightly sinuous between ovary and cirrus-sac and then straight and dorsal or sinistral to cirrus-sac. Metraterm thin, muscular, thickening slightly terminally into small sphincter-like structure, thickening slightly prior to sphincter into poorly defined chamber measuring 25–55 (40). Eggs few.

Excretory vesicle tubular, straight, often relatively dilate, extends anteriorly to about level of seminal vesicle and anterior extent of vitelline field. Excretory pore small, terminal.

3.2.3.1. Taxonomic summary

Type-host: Chaetodon quadrimaculatus Gray (Perciformes: Chaetodontidae), fourspot butterflyfish.

Type-locality: off O‘ahu, Hawai‘i.

Site of infection: intestine.

Other records: Chaetodon auriga Forsskål, threadfin butterflyfish, C. fremblii Bennett, bluestriped butterflyfish and C. multicinctus Garrett, pebbled butterflyfish, off O‘ahu (Pritchard,

1966); Zanclus cornutus (Linnaeus) (Perciformes: Zanclidae), Moorish idol (as Z. canescens), from

Hawaiian waters (Pritchard, 1966; Yamaguti, 1970).

Dubious records: Chaetodon rainfordi McCulloch, Rainford’s butterflyfish and C. plebius

Cuvier, blueblotch butterflyfish, off Heron Island, Great Barrier Reef, Australia (Bray & Cribb,

1989). Chaetodon lunula (Lacépède), racoon butterflyfish, from Seychelloise waters (Toman,

1992).

Second intermediate hosts: Porites compressa Dana (Anthozoa: Poritidae) and Por. lobata

Dana in Kāne‘ohe Bay, O‘ahu, as Plagioporus sp. (Cheng & Wong, 1974); Por. compressa in

Kāne‘ohe Bay (Aeby, 1998).

Prevalence: 6 of 6 (100%) C. quadrimaculatus and 4 of 16 (25%) C. multicinctus collected off O‘ahu (this study).

Material deposited: 8 voucher specimens, mounted laterally, comprising 6 from 2 C. quadrimaculatus off Hale‘iwa, Waialua, O‘ahu (21°36'N 158°7'W) (USNM 1480283 – 1480288) and 2 from 1 C. multicinctus off Kāne‘ohe Bay, O‘ahu (21°26'N 157°47'W) (USNM 1480289 –

1480290).

Representative DNA sequence data: 1 representative sequence deposited in GenBank for each of 4 identical partial 5.8S-ITS2-partial 28S rDNA (MH823948), 2 identical partial 28S rDNA

(MH823954) and 2 identical partial 18S rDNA (MH823959).

3.2.3.2. Remarks

Our specimens of P. stenometra are consistent with those of Pritchard (1966) and were recovered from two of the same host fishes, including the type-host, collected in the same region.

Pritchard (1966) originally described a canalicular seminal receptacle, but Bray and Cribb (1989) determined that the seminal receptacle is uterine after examining her specimens and specimens they identified as P. stenometra from chaetodontids collected on the Great Barrier Reef. It is unlikely that their specimens, or those of Toman (1992) from the Seychelles, represent P. stenometra (sensu stricto). Toman (1992) provided only a brief description, but his specimens are substantially larger than those examined here (2,040–2,880 µm versus 978–1,594 (1,320) µm long). Bray and Cribb

(1989) provided a detailed description with illustrations; in their specimens the body size is comparable with those examined here, but the pharynx is smaller than the oral sucker versus larger, the testes are contiguous and the anterior testis contiguous with the ovary versus all gonads usually separate, the caeca are consistently subequal versus usually equal, the excretory vesicle extends

12 almost to the level of the ovary or a little beyond versus always well beyond, and the eggs are substantially larger (63–70 (65) µm versus 50–55 (52) µm long).

Pritchard (1966), and later Yamaguti (1970), reported P. stenometra from Zanclus cornutus collected in Hawaiian waters. These are the only records of a species of Polypipapiliotrema from a fish other than a chaetodontid and, as far as we are aware, of any opecoelid from that fish. Zanclus cornutus is not usually considered a corallivore (Cole et al., 2008) but it has been observed feeding on hard corals (Powell et al., 2015), including on Porites lobata off the Galápagos (Feingold in

Enochs and Glynn, 2016). We have examined 13 Z. cornutus from French Polynesian waters and 34 on the Great Barrier Reef, and have not encountered any opecoelids. It is possible that Z. cornutus exhibits biogeographic dietary variation, whereby it preys relatively significantly upon Porites spp. in Hawaiian waters or, conversely, perhaps prevalence of Porites trematodiasis is relatively high in

Hawaiian waters (Aeby, 2006, 2007), such that incidental levels of predation by Z. cornutus are sufficient to enable readily detectable infections there. However, considering the diversity of species of Polypipapiliotrema discovered here, the close morphological similarity among these species and the importance of host specificity, we propose an alternative hypothesis: we predict that the specimens collected by Pritchard (1966) and Yamaguti (1970) from Z. cornutus represent a distinct species, one that switched from chaetodontids in Hawaiian waters and thus is likely endemic there. The specimens collected by Yamaguti (1970) from Z. cornutus are substantially larger than those from Hawaiian chaetodontids examined here (1,300–2,200 µm versus 978–1,594

(1,320) µm long), providing some evidence that two species might be involved. Recollection and sequencing of specimens from Hawaiian Z. cornutus is required to test this hypothesis.

3.2.4. Polypipapiliotrema heniochi Martin, Cutmore & Cribb n. sp. (Fig. 2 C, D)

Synonyms: Podocotyloides stenometra of Lucas et al. (2005); Podocotyloides stenometra of

Martin et al. (2018c).

Description [based on 15 lateral whole-mounts (Table 3)]. Body narrow, elongate, subcylindrical. Forebody short. Tegument smooth, about 7 thick. Oral sucker terminal, unspecialised, ellipsoidal with depth greater than length. Ventral sucker supported by relatively long, narrow peduncle, ellipsoidal with length greater than depth, larger than oral sucker.

Prepharynx short, prominent. Pharynx unspecialised, subspherical with depth usually slightly greater than length, usually slightly smaller than oral sucker. Oesophagus gently sinuous. Intestine bifurcates dorsal or immediately anterior to ventral sucker. Caeca blind, terminate well beyond testes, usually equal.

Testes 2, smooth, ellipsoidal with length greater than depth, similar in size with posterior testis usually slightly larger than anterior testis, tandem, ventral, separated by short gap of 6–157

(58). Cirrus-sac thin, elongate, tapering slightly distally, gently curved to sigmoid, extending beyond ventral sucker into hindbody. Internal seminal vesicle narrow, unipartite. Pars prostatica short, poorly differentiated from seminal vesicle. Ejaculatory duct long, thin, muscular, simple.

Genital atrium simple. Genital pore small, simple, ventral, medial or almost so, prebifurcal.

Ovary smooth, subspherical, sinistro-submedial, ventral, anterior to and separated by short gap [6–157 (58)] from anterior testis. Vitelline reservoir smaller than and antero-dorsal to ovary.

Vitellarium composed of small follicles restricted to hindbody, distributed laterally and dorsally from posterior extremity to about level of seminal vesicle and ventrally in post-testicular region, uninterrupted. Oviduct short. Oötype well-defined. Mehlis’ gland diffuse, inconspicuous. Uterus thin, entirely anterior to ovary, ascending ventrally in tight coils between ovary and cirrus-sac and then straight and dorsal or sinistral to cirrus-sac. Metraterm thin, muscular, simple, without chamber or sphincter. Eggs relatively numerous.

Excretory vesicle tubular, straight, often relatively dilate, extends anteriorly to about level of seminal vesicle and extent of vitelline field. Excretory pore small, terminal.

3.2.4.1. Taxonomic summary

Type-host: Heniochus chrysostomus Cuvier (Perciformes: Chaetodontidae), pennant bannerfish.

Type-locality: off Papetō'ai, Moorea, Society Islands, French Polynesia (17°29'S 149°53'W).

Site of infection: intestine.

Other host-locality combinations: Heniochus monoceros Cuvier, masked bannerfish, off

Tubuai (23°22’S 149°28’W), Austral Islands and off Toau (15°52'S 146°01'W), Tuamotu Islands;

H. chrysotomus off Heron Island, Great Barrier Reef, Australia (Lucas et al., 2005).

Prevalence: For H. chrysostomus, 8 of 29 (28%) off Moorea, 2 of 2 off Tubuai and 5 of 6

(83%) off Toau, with no infection detected from 1 off Rimatara in the Austral Islands, 4 from the

Mangareva group in the Gambier Islands and 5 off Fakarava in the Tuamotu Islands. For H. monoceros, 1 of 1 off Moorea, 1 of 1 off Tubuai and 1 of 2 (50%) off Toau, with no infections detected from 2 each off Raivavae and Rimatara in the Austral Islands, 1 from the Mangareva group in the Gambier Islands and 3 off Fakarava in the Tuamotu Islands.

Material deposited: Holotype and 9 paratypes, mounted laterally, from 3 H. chrysostomus off Moorea, Society Islands (MNHN HEL761–HEL770); 4 voucher specimens, mounted laterally, including 3 from 2 H. chrysostomus off Tubuai, Austral Islands (MNHN HEL771–HEL773) and 1 from H. chrysostomus off Toau, Tuamotu Islands (MNHN HEL774).

Representative DNA sequence: 9 identical partial 5.8S-ITS2-partial 28S rDNA sequences matching previously published sequence data available at GenBank [DQ083434 (Lucas et al., 2005) and MF926404 (Martin et al., 2018c)]; 1 and 5 generated from specimens recovered from H. chrysostomus and 1 and 2 from H. monoceros collected off Tubuai and Toau, respectively.

Previously published sequence data available in GenBank for partial 18S rDNA (MF926405;

Martin et al., 2018c) and partial 28S rDNA (MF926406; Martin et al., 2018c).

Zoobank registration: http://zoobank.org/urn:lsid:zoobank.org.act:099D8E1C-69F8-4F23-

A277-34CE14F8E99C

Etymology: This species is named after its definitive hosts, species of Heniochus, the bannerfishes, because it is the only species of Polypipapiliotrema known from fishes of that genus and, likewise, the only species not known from fishes of the genus Chaetodon.

3.2.4.2. Remarks

Polypipapiliotrema heniochi is larger than P. stenometra and, relative to body length, has a shorter forebody and pre-vitelline zone, and a more anteriorly situated ovary and genital pore (Fig.

3). The cirrus-sac is also relatively shorter in P. heniochi, with a relatively longer seminal vesicle and relatively shorter pars prostatica. The pharynx is, on average, slightly smaller than the oral sucker in P. heniochi, whereas it is larger than the oral sucker in P. stenometra. Finally, P. heniochi has the thickest tegument of all species of Polypipapiliotrema, lacks a sphincter and chamber associated with the metraterm as in P. stenometra, and has more numerous and slightly larger eggs than that species. Our sequence data for this species matches those published by Lucas et al. (2005) and Martin et al. (2018c). Therefore, their specimens, identified as P. stenometra, actually represent

P. heniochi. Thus, P. heniochi has a demonstrated distribution spanning from French Polynesia to the Great Barrier Reef, is the only species of Polypipapiliotrema known to infect species of

Heniochus and is also the only species which apparently does not exploit species of Chaetodon.

3.2.5. Polypipapiliotrema hadrometra Martin, Cutmore & Cribb n. sp. (Fig. 4 A, B)

Description [based on 7 lateral whole-mounts (Table 3)]. Body narrow, elongate, subcylindrical, relatively large. Forebody short. Tegument smooth, about 5 thick. Oral sucker terminal, unspecialised, ellipsoidal with length usually greater than depth. Ventral sucker supported by broad, short peduncle, ellipsoidal with length greater than depth, larger than oral sucker.

Prepharynx short, prominent. Pharynx unspecialised, roughly spherical, smaller than oral sucker.

Oesophagus gently sinuous, surround by sparse gland cells. Intestine bifurcates dorsal or immediately anterior to ventral sucker. Caeca blind, terminate well beyond testes, usually equal.

Testes 2, smooth, ellipsoidal with length greater than depth, similar in size with posterior testis usually slightly larger than anterior testis, tandem, ventral, separated by short gap of 8–87

(43). Cirrus-sac thin, elongate, sigmoid, extending beyond ventral sucker into hindbody. Internal seminal vesicle narrow, unipartite. Pars prostatica narrow, elongate, distinct from seminal vesicle.

Ejaculatory duct long, thin, muscular, simple. Genital atrium simple. Genital pore small, simple, ventral, medial or almost so, prebifurcal.

Ovary smooth, subspherical, medial, ventral, anterior to and separated by short gap [23–100

(62)] from anterior testis. Vitelline reservoir smaller than and antero-dorsal to ovary. Vitellarium composed of small follicles restricted to hindbody, distributed laterally and dorsally from posterior extremity to about level of seminal vesicle and ventrally in post-testicular region, often interrupted laterally and dorsally at level of each (n=3) or either testis (n=1). Oviduct short. Oötype well- defined. Mehlis’ gland diffuse, may be inconspicuous. Uterus thin, entirely anterior to ovary, ascending ventrally in tight coils between ovary and cirrus-sac and then straight and dorsal or sinistral to cirrus-sac. Metraterm well-developed, muscular, thickening into prominent sphincter- like structure terminally, thickening into large, well-defined, highly glandular chamber prior to sphincter measuring 52–89 (69). Eggs relatively numerous.

Excretory vesicle tubular, straight, often relatively dilate, extends anteriorly well beyond ovary almost to level of ventral sucker. Excretory pore small, terminal.

3.2.5.1. Taxonomic summary

Type-host: Chaetodon lunula (Lacépède), racoon butterflyfish.

Type-locality: off Maria, Austral Islands, French Polynesia (21°48'S 154°42'W).

Site of infection: intestine.

Other host-locality combinations: none.

Prevalence: 2 of 2 C. lunula collected off Maria, with no infections detected in 1 examined off Tubuai, 15 off Moorea, 10 from sites in the Tuamotus, 6 from sites in the Marquesas and 2 from the Mangareva group in the Gambiers.

Material deposited: Holotype and 6 paratypes, mounted laterally, from 2 C. lunula, off

Maria, Austral Islands (MNHN HEL775–HEL781).

Representative DNA sequence: 1 sequence representative deposited in GenBank for 2 identical replicates of each region: partial 5.8S-ITS2-partial 28S rDNA (MH823949), partial 28S rDNA (MH823955) and partial 18S rDNA (MH823960).

Zoobank registration: http://zoobank.org/urn:lsid:zoobank.org.act:52175A27-9039-4F7D-

96C8-F001624CC2DB

Etymology: The specific name hadrometra is from Greek, composed of the adjective hadros, meaning thick or strong, and metra, the uterus, because of the specialised thickenings of the metraterm which are especially prominent in this species. The composite name is treated as a noun in apposition.

3.2.5.2. Remarks

Polypipapiliotrema hadrometra is the largest species in its genus and is characterised by two prominent features of its metraterm: a large, thick-walled chamber surrounded by dense gland cells, and an obvious thickening into a sphincter-like structure immediately prior to the genital atrium.

Both these characters are present in P. stenometra, but each in a notably reduced form; the metraterm of P. heniochi has neither feature. The distribution of vitelline follicles in P. hadrometra is often interrupted laterally and dorsally at the level of each or either testis, as opposed to usually uninterrupted in P. stenometra and P. heniochi. Compared with both those species, the oesophagus of P. hadrometra is relatively shorter, its oral sucker is longer than deep versus deeper than long, its pharynx is smaller relative to the oral sucker, its sucker depth ratio is greater, and its eggs are larger

(Fig. 3). The cirrus-sac in P. hadrometra, relative to body length, is longer than in P. heniochi but

18 shorter than in P. stenometra and is more similar to the latter in its internal configuration, with a clearly distinct pars-prostatica; compared with P. heniochi, the pars-prostatica is relatively longer and the seminal vesicle relatively. Finally, P. hadrometra has a relatively shorter pre-bifurcal zone than P. stenometra and a longer pre-ovarian zone and shorter post-testicular zone than for P. heniochi.

3.2.6. Polypipapiliotrema ovatheculum Martin, Cutmore & Cribb n. sp. (Fig. 4 C, D)

Description [based on 2 lateral whole-mounts (Table 3)]. Body narrow, elongate, subcylindrical, relatively large. Forebody short. Tegument smooth, about 3 thick. Oral sucker terminal, unspecialised, roughly spherical. Ventral sucker supported by broad, short peduncle, ellipsoidal with length greater than depth, larger than oral sucker. Prepharynx short, prominent.

Pharynx unspecialised, ellipsoidal with length greater than depth, smaller than oral sucker.

Oesophagus gently sinuous, surround by sparse gland cells. Intestine bifurcates dorsal or immediately anterior to ventral sucker. Caeca blind, terminate well beyond testes, equal.

Testes smooth, ellipsoidal with length greater than depth, similar in size with posterior testis slightly larger than anterior testis, tandem, ventral, contiguous or separated by a short gap of 17.

Cirrus-sac thin, elongate, sigmoid, extending beyond ventral sucker into hindbody. Internal seminal vesicle narrow, unipartite. Pars prostatica narrow, elongate, distinct from seminal vesicle.

Ejaculatory duct long, thin, muscular, simple. Genital atrium simple. Genital pore small, simple, ventral, medial or almost so, prebifurcal.

Ovary smooth, subspherical, medial, ventral, anterior to and separated by short gap [29, 31] from anterior testis. Vitelline reservoir smaller than and antero-dorsal to ovary. Vitellarium composed of small follicles restricted to hindbody, distributed laterally and dorsally from posterior extremity to about level of seminal vesicle and ventrally in post-testicular region, uninterrupted.

Oviduct short. Oötype well-defined. Mehlis’ gland diffuse, may be inconspicuous. Uterus thin, entirely anterior to ovary, ascending ventrally in tight coils between ovary and cirrus-sac and then

19 straight and dorsal or sinistral to cirrus-sac. Metraterm well-developed, muscular, without terminal sphincter, provided with thickened chamber. Eggs relatively numerous.

Excretory vesicle tubular, straight, often relatively dilate, extends anteriorly well beyond ovary almost to level of ventral sucker. Excretory pore small, terminal.

3.2.6.1. Taxonomix summary

Type-host: Chaetodon flavirostris Günther, dusky butterflyfish.

Type-locality: off Raivavae, Austral Islands, French Polynesia (23°52'S 147°40'W).

Site of infection: intestine.

Other host-locality combinations: Chaetodon auriga Forsskål, threadfin butterflyfish, off

Tubuai (23°22’S 149°28’W), Austral Islands; C. flavirostris off Rimatara (22°39'S 152°46'W),

Austral Islands; C. mertensii Cuvier, atoll butterflyfish, off Raivavae.

Prevalence: 1 of 5 (20%) C. auriga off Tubuai; 2 of 4 (50%) and 1 of 3 (33%) C. flavirostris off Raivavae and Rimatara, respectively; 1 of 2 (50%) C. mertensii off Raivavae. No infections detected in 1 C. auriga each from Maria, Raivavae and Rurutu, and 16, 12, 10 and 3 from off

Moorea, the Tuamotus, the Gambiers and the Marquesas, respectively; 2, 1, 4 and 1 from off

Rurutu, off Maria, the Gambiers and off Moorea, respectively; 1, 3 and 1 C. mertensii from Tubai, the Gambiers and the Marquesas, respectively.

Material deposited: Holotype and 2 paratypes, 1 each from C. flavirostris and C. mertensii

(hologenophore) off Raivavae (MNHN HEL782–HEL783).

Representative DNA sequence: 1 representative sequence deposited in GenBank for each region: partial 5.8S-ITS2-partial 28S rDNA (MH823950) representative of 4 identical replicates, and partial 28S rDNA (MH823956) and partial 18S rDNA (MH823961), each representative of 2 identical replicates.

Zoobank registration: http://zoobank.org/urn:lsid:zoobank.org.act:B77A5974-ADD2-49CF-

909F-9240580A215E

Etymology: The specific name ovatheculum is composed of two Latin nouns, ova, eggs, and theca, a case, modified with the suffix -ulum to a diminutive form. The name is chosen because this species, like P. hadrometra, to which it is most similar, is characterised by a chamber in its metraterm, however, the worm is overall smaller and the pouch is less well-developed.

3.2.6.2. Remarks

Polypipapiliotrema ovatheculum closely resembles P. hadrometra but is smaller, usually has a continuous versus often interrupted vitelline field, and its metraterm has no terminal sphincter and a small, poorly developed versus prominent proximal chamber. Morphometric comparison is limited due to sample size; however, P. ovatheculum has the largest eggs of any species of

Polypipapiliotrema and, compared with P. hadrometra, it appears that its cirrus-sac is relatively longer and its pharynx length to depth ratio is greater (Fig. 3). Similar to P. hadrometra, it is known only from the Austral Islands.

3.2.7. Polypipapiliotrema citerovarium Martin, Cutmore & Cribb n. sp. (Fig. 5 A-D)

Description [based on 22 lateral and 5 ventral whole-mounts (Table 3) from fishes collected off the Austral and Gambier Islands]. Body narrow, elongate, subcylindrical. Forebody short.

Tegument smooth, about 3 thick. Oral sucker terminal, unspecialised, ellipsoidal with depth greater than length. Ventral sucker supported by broad, short peduncle, ellipsoidal with length greater than depth, larger than oral sucker. Prepharynx short, prominent. Pharynx unspecialised, subspherical with length usually slightly greater than depth, larger than oral sucker. Oesophagus gently sinuous.

Intestine bifurcates dorsal or immediately anterior to ventral sucker. Caeca blind, terminate well beyond testes, usually subequal.

Testes 2, smooth, ellipsoidal with length greater than depth, similar in size, tandem to slightly diagonal, ventral, almost contiguous or almost so (n=1, gap of 14). Cirrus-sac thin, elongate, tapering slightly distally, gently curved to sigmoid, extending beyond ventral sucker into

21 hindbody. Internal seminal vesicle narrow, elongate, unipartite. Pars prostatica short, poorly differentiated from seminal vesicle. Ejaculatory long, thin, muscular, simple. Genital atrium simple.

Genital pore small, simple, ventral, medial or almost so, prebifurcal.

Ovary smooth, subspherical, medial, ventral, anterior to and contiguous with anterior testis.

Vitelline reservoir smaller than and antero-dorsal to ovary. Vitellarium composed of poorly differentiated to irregularly dendritic follicles scattered throughout hindbody, distributed laterally and dorsally from posterior extremity to about level of seminal vesicle and ventrally in post- testicular region, uninterrupted. Oviduct short. Oötype well-defined. Mehlis’ gland diffuse, inconspicuous. Uterus thin, entirely anterior to ovary, ventral and tightly sinuous between ovary and cirrus-sac and then straight and dorsal or sinistral to cirrus-sac. Metraterm thin, muscular, simple, without sphincter or chamber. Eggs few.

Excretory vesicle tubular, straight, often relatively dilate, extends anteriorly well beyond ovary to level of seminal vesicle. Excretory pore small, terminal.

3.2.7.1. Taxonomic summary.

Type-host: Chaetodon quadrimaculatus Gray, fourspot butterflyfish.

Type-locality: off Raivavae, Austral Islands, French Polynesia (23°52'S 147°40'W).

Site of infection: intestine.

Other host-locality combinations: Chaetodon citrinellus Cuvier, citron butterflyfish, C. mertensii and C. reticulatus Cuvier, black butterflyfish, off Raivavae; C. pelewensis Kner, dot-and- dash butterflyfish and C. quadrimaculatus off Maria (21°48'S 154°42'W), Austral Islands; C. citrinellus, C. mertensii and C. quadrimaculatus off Tubuai (23°22’S 149°28’W), Austral Islands;

C. quadrimaculatus and C. pelewensis from the Mangareva island group (23°06'S 134°58'W),

Gambier Islands.

Prevalence: For C. quadrimaculatus, 2 of 3 (67%) off Maria, 2 of 3 (67%) off Raivavae, 1 of 1 off Tubuai and 3 of 3 from the Mangareva group in the Gambier Islands, with no infections

22 detected in 32, 6 and 3 from off Moorea, Toau in the Tuamotu Islands and the Marquesas Islands, respectively. For C. citrinellus, 1 of 1 off Raivavae and 2 of 5 (40%) off Tubuai, with no infections detected in 66, 15 and 8 off Moorea, the Tuamotu Islands and the Marquesas Islands, respectively.

For C. mertensii, 2 of 2 off Raivavae and 1 of 1 off Tubuai, with no infections detected in 3 and 1 from the Gambier Islands and the Marquesas Islands, respectively. For C. pelewensis, 2 of 5 (40%) off Maria and 4 of 6 (67%) from the Mangareva group in the Gambier Islands, with no infections detected in 19, 9, 5, 4 and 1 from off Moorea, the Tuamotu Islands, the Marquesas Islands,

Raivavae and Rimatara, respectively. For C. reticulatus, 3 of 6 off Raivavae, with no infections detected in 22, 11, 3, 2 and 1 from off Moorea, the Tuamotu Islands, the Gambier Islands, the

Marquesas Islands and Rimatara.

Material deposited: Holotype and 9 paratypes, mounted laterally, from 1 C. quadrimaculatus off Raivavae (MNHN HEL785– HEL794); 10 vouchers, 7, including 4 lateral and

3 ventral mounts, from 3 C. quadrimaculatus, and 3, including 1 lateral and 2 ventral mounts, from

3 C. pelewensis, from the Mangareva island group, Gambier Islands (MNHN HEL795– HEL804).

Representative DNA sequence: Representative sequences deposited in GenBank for 3 distinct genotypes: partial 5.8S-ITS2-partial 28S rDNA representative of 6, 4 and 5 identical replicates for the Raivavae/Maria (MH823953), Tubuai (MH823952) and Gambiers (MH823951) genotypes, respectively; partial 28S rDNA representative of 3 and 2 identical replicates for the

Raivavae/Maria (MH823958) and Tubuai (MH823957) genotypes, respectively; partial 18S rDNA representative, 2 and 1 replicates for the Raivavae/Maria (MH823963) and Tubuai (MH823962) genotypes, respectively. No 18S or 28S data generated for the Gambiers genotype.

Zoobank registration: http://zoobank.org/urn:lsid:zoobank.org.act:97AA3FB2-3A3F-4899-

B2B1-8F8DFCE34B57

Etymology: The specific name citerovarium is composed from the Latin adjective citer, meaning lying near, and ovarium, the ovary, because the most conspicuous feature of this species,

23 relative to other species of Polypipapiliotrema, is the consistent absence of space between the ovary and the anterior testis. The name is treated as a noun in apposition.

3.2.7.2. Remarks

Polypipapiliotrema citerovarium is most readily distinguished from the other French

Polynesian species by contiguity of the testes and of the ovary with the anterior testis. In P. ovatheculum the gonads are close such that the testes may be contiguous or almost so, but there is always a small but distinct gap between the ovary and the anterior testis. In P. hadrometra and P. heniochi, the gonads are usually distinctly separate. Polypipapiliotrema citerovarium is most similar to P. stenometra, for which Pritchard (1966) described the testes as separate or contiguous.

Both species are small, have a larger pharynx than oral sucker, and have relatively few eggs compared with the other three species from French Polynesia. Polypipapiliotrema citerovarium can be distinguished from P. stenometra by the configuration of its terminal genitalia, which is more similar to that of P. heniochi; the pars-prostatica is poorly distinguished from the seminal vesicle and the metraterm lacks both the proximal chamber and distal sphincter. Morphometrically, P. citerovarium differs from P. stenometra by several measurements taken relative to body length: its longer forebody, longer pre-genital pore zone, shorter cirrus-sac, slightly shorter pre-testicular zone, longer post-caeca zone, as well as its slightly larger eggs (Fig. 3). Polypipapiliotrema citerovarium shares one definitive host species, C. quadrimaculatus, with P. stenometra (in allopatry), and another, C. mertensii, with P. flavirostris (in sympatry) at Raivavae.

We consider P. citerovarium to include three genotypes relating to biogeographic variants

(Fig. 6). The type genotype is from specimens collected off Raivavae and Maria, in the Austral

Islands. This genotype differs consistently by a single pyrimidine transition (cytosine versus thymine) in the ITS2 region to specimens from fishes collected off Tubuai, also of the Austral

Islands but located between Raivavae and Maria (Fig. 1). The third genotype relates to specimens from fishes collected in the Gambier Islands; it is consistent with the Raivavae/Maria genotype at

24 the variable site mentioned above, but differs from both that genotype and the Tubuai genotype by a purine transition (guanine versus adenosine) at another site in the ITS2 region (i.e. it differs to the

Raivavae/Maria genotype by one base position and to the Tubuai genotype by two base positions).

It was not possible to generate 28S or 18S data from specimens collected in the Gambier Islands, but the Tubuai genotype was identical to the Raivavae/Maria genotype in the 18S marker and differed at three consistent base-positions in the 28S marker. By comparison, minimum interspecific differences for the ITS2, 28S and 18S among what are here interpreted as distinct species of

Polypipapiliotrema are four, eight and 2 bp, respectively (Fig. 6, Table 4).

Insufficient material was collected at Tubuai, but morphometric comparison between specimens from the Gambier Islands and Raivavae + Maria was possible. Qualitatively, specimens from these sites are indistinguishable, but some morphometric differences were detected (Fig. 7).

Most compellingly, the distance between the ovary and the ventral sucker is relatively shorter in specimens from Raivavae and Maria and the seminal vesicle is shorter and ejaculatory duct therefore longer relative to the length of the cirrus-sac. Additionally, the forebody is longer in specimens from Raivavae/Maria, the oral sucker and pharynx are more similar and the pharynx length to depth ratio is greater. Although these are detectable, the effect sizes are small, with considerable overlap. These comparisons assume that all specimens collected at each site represent the respective genotype discovered there. In a similar situation, we previously (Martin et al., 2018c) considered two morphologically indistinguishable genetic variants of Podocotyloides species as distinct, but in that study a difference was detected in all three ribosomal markers and the magnitude of the difference for each marker between the two cryptic species was comparable to differences among other morphologically distinctive species of Podocotyloides.

3.3. Phylogenetic analyses

Phylograms produced through Bayesian inference (Fig. 8) and maximum likelihood (not figured) analyses of the concatenated 28S + 18S rDNA dataset were essentially consistent in

25 topology and also consistent with previous phylogenetic assessments of the Opecoelidae (Martin et al., 2018c). Species of Polypipapiliotrema formed a strongly supported monophyletic and phylogenetically distinct clade, resolving among the large group of taxa from marine fishes which have typically been consigned to the Plagioporinae (sensu lato). As per previous investigations, the

Plagioporinae (sensu stricto) should now be considered restricted to freshwater and possibly deep- sea species (Fayton and Andres, 2016; Martin et al., 2017, 2018a). So far, the large marine plagioporine (sensu lato) group has been found to comprise three genetically distinct clades, one of which is recognised as the Opistholebetinae Fukui, 1929 (see Martin et al., 2018b), whereas the other two, referred to as marine clade B and marine clade C, are currently without adequate subfamilial designation. Sequence data for species of Polypipapiliotrema did not resolve to any of these clades but fell as sister to Pseudopycnadena tendu Bray & Justine, 2007. That species also requires reassessment, because it is only distantly related to Pseudopycnadena fischthali Saad-Fares

& Maillard, 1986, the type-species, which resolved among the Opistholebetinae.

4. Discussion

Justification for recognition of the Polypipapiliotrematinae is based on combined evidence from phylogeny, morphology and the unique incorporation of coral second intermediate hosts into the life-cycle. Polypipapiliotrema cannot be assigned to the Plagioporinae, the original subfamilial designation of its type species, because its constituent taxa are phylogenetically distant from those of Plagioporus Stafford, 1904, the type genus. Furthermore, the Plagioporinae is defined for species with a canalicular seminal receptacle, whereas species of Polypipapiliotrema are the only known taxa among the large marine plagioporine (sensu lato) clade to lack this feature. The Stenakrinae

Yamaguti, 1970, a subfamily for which no sequence data are yet available, is defined for species without a canalicular seminal receptacle, but species of Polypipapiliotrema are not morphologically similar to any stenakrine generic concepts and, perhaps most significantly, it is unlikely that any stenakrine taxa, or indeed any other known opecoelids, exploit anthozoans at any stage in their life-

26 cycle. Finally, although species of Polypipapiliotrema resolved most closely to Pseudopycnadena tendu, we do not think that species should be considered under the new subfamily. It was separated from species of Polypipapiliotrema by long branch lengths, has a canalicular seminal receptacle and shares no significant morphological features with those species. Its only known definitive host, the yellow-spotted triggerfish Pseudobalistes fuscus (Bloch & Schneider) (Tetraodontiformes:

Balistidae), is known to graze on corals but has a broad diet including many benthic invertebrates

(Froese and Pauly, 2018) and is unrelated to chaetodontids.

Recognition of the Polypipapiliotrematinae contributes significant evidence suggesting that divergence of some major lineages within the Opecoelidae (i.e. at the subfamilial level) may have been strongly driven by second intermediate host switch events. Arthropods, especially crustaceans, are the second intermediate hosts for most known opecoelid life-cycles, including for the

Helicometrinae Bray, Cribb, Littlewood & Waeschenbach, 2016 (Meenakshi et al., 1993; Jousson et al., 1999; Leiva et al., 2017), the Opecoelinae Ozaki, 1925 (e.g. Cribb, 1985; Jousson and Bartoli,

2000; Yoshida and Urabe, 2005), potentially the deep-sea taxa (Thompson and Margolis, 1987;

Blend and Dronen, 2015b), and some (Schell, 1975; Hendrix, 1978), but not all (Sinitsin, 1931;

Schell, 1976; Yano and Urabe, 2017) freshwater plagioporines (sensu stricto). In contrast, only opistholebetines are known to use gastropod and echinoid second intermediate hosts (Martin et al.,

2018b) and the few life-cycles known from the major lineage referred to as the ‘Plagioporinae

(sensu lato) clade B’ exploit fishes as second intermediate hosts (McCoy, 1929, 1930; Downie, A.

2011. Patterns of trophic transmission of digenean trematodes through fishes of the Great Barrier

Reef: host specificity or hoping for the best? Ph.D thesis (unpublished), University of Queensland,

Brisbane, Australia.). In this context, it is significant that species of Polypipapiliotrema are the only opecoelids known to exploit corals as second intermediate hosts. We predict that future subfamilial classification refinement among the Opecoelidae, especially among the major marine Plagioporinae

(sensu lato) clade, will increasingly align with distinctions in life-history, as the breadth of opecoelid taxa for which both these and sequence data become increasingly available.

The distinction of the Polypipapiliotrematinae and potentially other opecoelid clades on the basis of life-cycle is consistent with a paradigm of trematode evolution whereby significant second intermediate host switch events appear to have underpinned the rise of major lineages. Comparable examples include the Bucephalidae Poche, 1907 and Opisthorchioidea Braun, 1901, each of which apparently arose following adoption of fish second intermediate hosts (Cribb et al., 2003).

Examples at finer resolutions include within the Echinostomatidae, in which there is a major dichotomy between taxa which use gastropod versus fish second intermediate hosts (Tkach et al.,

2016), and within the Diplostomidae Poirier, 1886, in which an intermediate host switch from fishes to amphibians appears to have enabled a definitive host switch from birds to mammals, giving rise to the Alariinae Hall & Wigdor, 1918 (Shoop, 1989; Blasco-Costa and Locke, 2017). The theory also extends to some groups that appear to have secondarily adopted a two-host life-cycle in which metacercariae encyst in the open, including twice among the Echinostomatoidea Looss, 1889, for the Fasciolidae Railliet, 1895 and Philophhalmidae Looss, 1889 (see Tkach et al., 2016), and for some lineages which exploit herbivorous fishes; apparently twice among the Lepocreadiodea, for the Gyliauchenidae Fukui, 1929 + Enenteridae Yamaguti, 1958 and the Gorgocephalidae Manter,

1966 (Bray and Cribb, 2012; Huston et al., 2016), and probably also for the Haploporoidea Nicoll,

1914 and Haplosplanchnata Olson, Cribb, Tkach, Bray & Littlewood, 2003 (Olson et al., 2003).

Thus, although life-cycle studies for most groups, including the Opecoelidae, are almost always limited, identification of key host switch events are clearly important for understanding the evolutionary history of many lineages.

Our investigation has uncovered a complex of species exploiting chaetodontids in French

Polynesia. Previous efforts to document the trematode fauna of French Polynesian fishes are few and have tended to treat the area, spanning over 2,000 km, as a single biogeographic unit, but corals and chaetodontid community composition, and chaetodontid feeding preferences, may vary at regional or even local scales (Cox, 1994; Berumen and Pratchett, 2006). Our findings provide a clear example of heterogeneity in the trematode fauna within French Polynesia. Moorea was best

28 sampled, but yielded only a single host-parasite combination, whereas all four species encountered in French Polynesia were detected in the Austral Islands with comparatively modest sampling effort. Similarly, the Marquesas Islands, which are characterised by scarce coral cover but dominated by species of Porites (Cabioch et al., 2011), are reasonably well-represented in our dataset but no infections of species of Polypipapiliotrema were detected; potentially none occur there. Patterns of host specificity and biogeographic distribution also varied among species. Three of the four species encountered in French Polynesia occurred in multiple definitive host species, but only one chaetodontid, C. mertensii, was host to more than a single species of Polypipapiliotrema there; additionally, C. quadrimaculatus is host to P. citerovarium in French Polynesia and P. stenometra in Hawai‘i. Polypipapiliotrema citerovarium was the only species encountered in the

Gambier Islands and, similarly, only P. heniochi was encountered in fishes from the Tuamotu

Islands and off Moorea. However, P. citerovarium exhibited consistent genetic variation between the Gambier and Austral Island groups and even among islands within the latter, whereas data generated for P. heniochi were identical between French Polynesia and the Great Barrier Reef, a distance of some 6,000 km. Several previous studies have reported lower diversity and prevalence of marine trematodes from French Polynesia relative to elsewhere in the Indo-West Pacific (Rigby et al., 1999; Cribb et al., 2014a), including from chaetodontids (McNamara et al., 2012; Díaz et al.,

2013). Thus, given the observed patterns of host specificity and biogeography here, and the ubiquity and richness of chaetodontids on tropical coral reefs in the Indo-Pacific, we predict that

Polypipapiliotrema represents a relatively large complex comprising both regional endemics and species with broad distributions.

Most chaetodontids incorporate corals at least to some extent into their diets and it is apparent from the patterns of host usage detected here that even chaetodontids considered to exhibit low levels of corallivory encounter species of Polypipapiliotrema with sufficient frequency to act as detectable hosts. For example, corals constitute less than 5% of the diet of C. auriga on the northern

Great Barrier Reef (Pratchett, 2005) but it is host to P. ovatheculum and P. stenometra (Pritchard,

1966), and H. monoceros feeds mainly on polychaetes in Japanese waters (Sano, 1989) but is host to P. heniochi. Similarly, Zanclus cornutus is not usually considered corallivorous but is reported to host P. stenometra (Pritchard, 1966; Yamaguti, 1970). Conversely, it is equally intriguing to consider the conspicuous absence of infection in certain chaetodontids examined. Most of the obligate corallivores belong to Chaetodon clade 3, but most fishes found here to host species of

Polypipapiliotrema belong to clades 2 and 4. In particular, C. lunulatus Quoy & Gaimard and C. ornatissimus Cuvier belong to clade 3 and were well represented in our dataset. Both species are known to feed at least to some extent on species of Porites (Cole et al., 2008) and in French

Polynesia these corals reportedly constitute the most important prey for C. lunulatus (Pratchett,

2014). However, no infections of species of Polypipapiliotrema were detected in these fishes.

Where a large host fauna such as the Chaetodontidae are readily available, it is typical for individual trematode species within a complex to infect only a proportion of the host species available (Cribb et al., 2014b), but this phenomenon is generally considered in terms of resource partitioning; the absence of any species within a complex from apparently appropriate and available hosts such as C. lunulatus or C. ornatissimus in French Polynesia is difficult to explain.

The apparently discrepant use of second intermediate host species for species of

Polypipapiliotrema is also striking. Suspected trematodiasis in corals has been reported only in species of Porites (Willis et al., 2001; Yamashiro, 2004; Aeby, 2006; Aeby, 2007; Vera and Banks,

2009; Aeby et al., 2015). However, among corals, species of Porites are comparatively poor food sources and most corallivorous chaetodontids prefer species of other genera (e.g. Acropora Oken

(Acroporidae), Montipora Blainville (Acroporidae) and/or Pocillopora Lamarck (Pocilloporidae))

(Gochfeld, 2004; Berumen and Pratchett, 2006; Cole et al., 2008; Nagelkerken et al., 2009;

Pratchett, 2014); sometimes species of Porites are even reported to be actively avoided (Pratchett,

2005; Cole et al., 2008). For corallivorous butterflyfishes, prey quality is affected by the interaction between meal size, nutrient content, handling time and the cost of enduring coral defences. Species of Porites respond to grazing by withdrawing polyps and, over longer temporal scales, increasing

30 local nematocyst density (Gochfeld, 2004). Infection by Polypipapiliotrema metacercariae facilitates parasite transmission by increasing attractiveness of the infected polyp to potential definitive hosts in three ways: polyp growth and thus meal size increases, coral defences are circumvented because the swollen polyp is unable to retract, and polyp vulnerability is advertised via the pink discolouration (Aeby, 2002). However, compared with other corals, species of Porites consistently develop pink discolouration in response to irritation, regardless of causative agent

(Work et al., 2014). Thus, we speculate that the apparent specialisation in species of

Polypipapiliotrema for species of Porites was driven by the opportunity to exploit, perhaps exaggerate, the existing conspicuous host immune response, and is not hindered but rather facilitated by the poor meal quality offered by these corals. This is because the difference in quality and accessibility between infected versus uninfected polyps might be substantially greater for species of Porites than for other corals such that exploitation of these species maximises transmission.

Porites trematodiasis has been reported from across the Indo-Pacific (Willis et al., 2001;

Yamashiro, 2004; Aeby, 2006, 2007; Vera and Banks, 2009; Aeby et al., 2015), but it is not clear whether all reported cases can be presumed to be caused by infection with species of

Polypapipiliotrema. First, outside of Hawaiian waters, only in one instance, in New Caledonia

(Work et al., 2014; Aeby et al., 2015), has a case of Porites trematodiasis been conclusively linked to trematode infection. Second, a similar life-cycle to that known for P. stenometra has been predicted by Díaz et al. (2013, 2015) for species relating to a complex of species of Paradiscogaster

Yamaguti, 1934 (Faustulidae Poche, 1926). Correct identification of a pathogen may be important for understanding the disease it causes and, similarly, appreciation of the diversity among pathogens causing the same (or apparently the same) disease improves the capacity to explain differences among observations. A lack of appreciation for this diversity would unknowingly undermine conclusions drawn from any study investigating trematodiasis in corals due to a failure to consider that observed differences (e.g. between coral species, between reefs or biogeographical areas,

31 between experimental treatments, or between studies) might be partially explained by nuances of the different parasite species involved. In the case of Polypipapiliotrema, none of the species encountered in this study has been found in both Hawaiian and French Polynesian fishes. Therefore, it is most likely that reports of P. stenometra from outside Hawai‘i, either in fishes or as trematodiasis in corals, actually represent different species; clearly the richness of this group is yet to be realised.

Acknowledgements

We thank Matthew Ross for procuring fishes in Hawai‘i; staff of the School of Ocean and

Earth Science and Technology, University of Hawai‘i research station at Moko o Lo‘e; Dr Serge

Planes and staff of the CRIOBE research station at Moorea French Polynesia; and two anonymous reviewers for their time, comments and excellent advice. This project was funded principally by a

PADI Foundation (USA) research grant awarded to SBM. Collection of specimens from Hawai‘i was enabled by a travel grant awarded to SBM by the School of Biological Sciences, the University of Queensland, Australia. Collection efforts in French Polynesia have been funded variously by

BioCode, France, the Coral Spot programme (supported by the French Polynesian and French governments), the Pakaihi i te Moana expedition [supported by Agence des Aires Marines

Protégées, France (AAMP) and the French Polynesian and French governments], the Khaled bin

Sultan Living Oceans Foundation, USA, and the University of Queensland, Australia. SBM is supported by a Holsworth Wildlife Research Endowment administered by the Ecological Society of

Australia, and a PhD scholarship provided through the Australian Government’s Australian

Biological Resources Study (ABRS) National Taxonomy Research Grant Programme (NTRGP).

References

Aeby, G.S., 1998. A digenean metacercaria from the reef coral, Porites compressa, experimentally

identified as Podocotyloides stenometra. J. Parasitol. 84, 1259–1261.

Aeby, G.S., 2002. Trade-offs for the butterflyfish, Chaetodon multicinctus, when feeding on coral

prey infected with trematode metacercariae. Behav. Ecol. Sociobiol. 52, 158–163.

Aeby, G.S., 2006. Baseline levels of coral disease in the Northwestern Hawaiian Islands. Atoll Res.

Bull. 543, 471–488.

Aeby, G.S., 2007. Spatial and temporal patterns of Porites trematodiasis on the reefs of Kaneohe

Bay, Oahu, Hawaii. Bull. Mar. Sci. 80, 209–218.

Aeby, G.S., Tribollet, A., Lasne, G., Work, T.M. 2015. Assessing threats from coral and crustose

coralline algae disease on the reefs of New Caledonia. Mar. Freshwater Res. 67, 455–465.

Andres, M.J., Pulis, E.E., Overstreet, R.M., 2014a. New genus of opecoelid trematode from

Pristipomoides aquilonaris (Perciformes: Lutjanidae) and its phylogenetic affinity within

the family Opecoelidae. Folia Parasitol. 61, 223–230.

Andres, M.J., Ray, C.L., Pulis, E.E., Curran, S.S., Overstreet, R.M., 2014b. Molecular

characterization of two opecoelid trematodes from fishes in the Gulf of Mexico, with a

description of a new species of Helicometra. Acta Parasitol. 59, 405–412.

Antar, R., Georgieva, S., Gargouri, L., Kostadinova, A., 2015. Molecular evidence for the existence

of species complexes within Macvicaria Gibson & Bray, 1982 (Digenea: Opecoelidae) in

the western Mediterranean, with descriptions of two new species. Syst. Parasitol. 91, 211–

229.

Bellwood, D.R., Klanten, S., Cowman, P.F., Pratchett, M.S., Konow, N., van Herwerden, L., 2010.

Evolutionary history of the butterflyfishes (f: Chaetodontidae) and the rise of coral feeding

fishes. J. Evol. Biol. 23, 335–349.

Berumen, M.L., Pratchett, M.S., 2006. Recovery without resilience: persistent disturbance and long-

term shifts in the structure of fish and coral communities at Tiahura Reef, Moorea. Coral

Reefs 25, 647–653.

Berumen, M.L., Pratchett, M.S., Goodman, B.A., 2011. Relative gut lengths of coral reef

butterflyfishes (Pisces: Chaetodontidae). Coral Reefs 30, 1005–1010.

Blasco-Costa, I., S.A., Locke, 2017. Life history, systematics and evolution of the Diplostomoidea

Poirier, 1886: progress, promises and challenges emerging from molecular studies. Adv.

Parasitol. 98, 167–225.

Blend, C.K., Dronen, N.O., 2015a. A review of the genus Helicometra Odhner, 1902 (Digenea:

Opecoelidae: Plagioporinae) with a key to species including Helicometra overstreeti n. sp.

from the cuck-eel Luciobrotula corethromycter Cohen, 1964 (Ophidiiformes: Ophidiidae)

from the Gulf of Mexico. Mar. Biodivers. 45, 183–270.

Blend, C.K., Dronen, N.O., 2015b. Description of a new species of Podocotyle Dujardin, 1845

(Digenea: Opecoelidae: Plagioporinae) from the cusk-eel, Luciobrotula corethromycter

Cohen, 1964 (Ophidiiformes: Ophidiidae), from the Gulf of Mexico and Caribbean Sea..

Acta Parasitol. 60, 234–243.

Born-Torrijos, A., Kostadinova, A., Raga, J.A., Holzer, A.S., 2012. Molecular and morphological

identification of larval opecoelids (Digenea: Opecoelidae) parasitising prosobranch snails in

a Western Mediterranean lagoon. Parasitol. Int. 61, 450–460.

Bouchon-Navaro, Y., 1986. Partitioning of food and space resources by chaetodontid fishes on

coral reefs. J. Exp. Mar. Bio. Ecol. 103, 21–40.

Bray, R.A., Cribb, T.H., 1989. Digeneans of the family Opecoelidae Ozaki, 1925 from the southern

Great Barrier Reef, including a new genus and three new species. J. Nat. Hist. 23, 429–473.

Bray, R.A., Cribb, T.H., 2012. Reorganisation of the superfamily Lepocreaadioidea Odhner, 1905

based on an inferred molecular phylogeny. Syst. Parasitol. 83, 169–177.

Bray, R.A., Cribb, T.H., Littlewood, D.T.J., Waeschenbach, A., 2016. The molecular phylogeny of

the digenean family Opecoelidae Ozaki, 1925 and the value of morphological characters,

with the erection of a new subfamily. Folia Parasitol. 63, 1–11.

Bray, R.A., Justine, J.-L., 2009. Neolebouria blatta n. sp. (Digenea: Opecoelidae) from

Pristipomoides argyrogrammicus (Valenciennes) and Etelis carbunculus Cuvier

(Perciformes: Lutjanidae) off New Caledonia. Syst. Parasitol. 74, 161–167.

Bray, R.A., Justine, J.-L., 2012. Further reports of Acanthocolpidae Lühe, 1906 (Digenea) from

fishes off New Caledonia, with descriptions of two new species. Syst. Parasitol. 83, 39–50.

Bray, R.A., Waeschenbach, A., Dyal, P., 2012. New digeneans (Opecoelidae) from hydrothermal

vent fishes in the south eastern Pacific Ocean, including one new genus and five new

species. Zootaxa 3768, 73–87.

Bray, R.A., Webster, B.L., Bartoli, P., Littlewood, D.T.J., 2005. A molecular study of the

Acanthocolpidae (Digenea). Acta Parasitol. 50, 281–291.

Cabioch, G., Wallace, C.C., McCulloch, M.T., Zibrowius, H., Laboute, P., Richer de Forges, B.,

2011. Disappearance of Acropora from the Marquesas (French Polynesia) during the last

deglacial period. Coral Reefs 30, 1101–1105.

Cheng, T.C., Wong, A.K., 1974. Chemical, histochemical, and histopathological studies on corals,

Porites spp., parasitized by trematode metacercariae. J. of Invertebr. Pathol. 23, 303–317.

Cole, A.J., Pratchett, M.S., Jones, G.P., 2008. Diversity and functional importance of coral-feeding

fishes on tropical coral reefs. Fish Fish. 9, 1–22.

Constenla, M., Carrassón, M., Moyà, C.M., Fernàndez-Chacón, A., Padrós, F., Repullés-Albelda,

A., Montero, F.E., 2011. Parasitisation by Bathycreadium elongatum (Digenea,

Opecoelidae) in pyloric caeca of Trachyrincus scabrus (Teleostei, Macrouridae). Dis.

Aquat. Organ. 96, 239–247.

Cox, E.F., 1994. Resource use by corallivorous butterflyfishes (family Chaetodontidae) in Hawaii.

Bull. Mar. Sci. 54, 535–545.

Cribb., T.H., 1985. The life cycle and biology of Opecoelus variabilis sp. nov. (Digenea:

Opecoelidae). Aust. J. Zool. 33, 715–728.

Cribb., T.H., Bray, R.A., 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–

Cribb, T.H., Bray, R.A., Littlewood, D.T.J., Pichelin, S., Herniou, E.A., 2001. The Digenea. In:

Littlewood, D.T.J., Bray R.A. (Eds), Interrelationships of the Platyhelminthes. Taylor and

Francis, London, pp. 168–185.

Cribb, T.H., 2005. Family Opecoelidae Ozaki, 1925. In: Jones, A., Bray, R.A., Gibson, D.I., (Eds),

Keys to the Trematoda. Vol. 2. CABI Publishing and The Natural History Museum,

Wallingford, UK, pp. 443–531.

Cribb, T.H., Adlard, R.D., Bray, R.A., Sasal, P., Cutmore, S.C., 2014a. Biogeography of tropical

Indo-West Pacific parasites: a cryptic species of Transversotrema and evidence for rarity of

Transversotrematidae (Trematoda) in French Polynesia. Parasitol. Int. 63, 285–294.

Cribb, T.H., Bott, N.J., Bray, R.A., McNamara, M.K.A., Miller, T.L., Nolan, M.J., Cutmore, S.C.,

2014b. Trematodes of the Great Barrier Reef, Australia: emerging patterns of diversity and

richness in coral reef fishes. Int. J. Parasitol. 44, 929–939.

Cribb, T.H., Bray, R.A., Olson, P.D., Littlewood, D.T.J., 2003. Life cycle evolution in the Digenea:

a new perspective from phylogeny. Adv. Parasitol. 54, 197–254.

Curran, S.S., Overstreet, R.M., Tkach, V.V., 2007. Phylogenetic affinities of Plagiocirrus Van

Cleave and Mueller, 1932 with the description of a new species from the Pascagoula River,

Mississippi. J. Parasitol. 93, 1452–1458.

Díaz, P.E., Bray, R.A., Cribb, T.H., 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.

Díaz, P.E., Bray, R.A., Cutmore, S.C., Ward, S., Cribb, T.H., 2015. A complex of species related to

Paradiscogaster glebulae (Digenea: Faustulidae) in chaetodontid fishes (Teleostei:

Perciformes) of the Great Barrier Reef. Parasitol. Int. 64, 421–428.

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high

throughput. Nucleic Acids Res. 32, 1792–1797.

Enochs, I.C., Glynn, P.W., 2017. Corallivory in the Eastern Pacific. In: Glynn, P.W., Marcello, D.,

Enochs, I.C. (Eds), Coral Reefs of the World. Vol. 8. Coral Reefs of the Eastern Tropical

Pacific. Springer, Dordrecht, The Netherlands, pp. 315–337.

Fayton, T.J., Andres, M.J., 2016. New species of Plagioporus Stafford, 1904 (Digenea:

Opecoelidae) from California, with an amendment of the genus and a phylogeny of

freshwater plagioporines of the Holarctic. Syst. Parasitol. 93, 731–748.

Fayton, T.J., Choudhury, A., McAllister, C.T., Robison, H.W., 2017. Three new species of

Plagioporus Stafford, 1904 from darters (Perciformes: Percidae), with a redescription of

Plagioporus boleosomi (Pearse, 1924) Peters, 1957. Syst. Parasitol. 94, 159–182.

Fayton, T.J., McAllister, C.T., Robison, H.W., Connior, M.B., 2018. Two new species of

Plagioporus (Digenea: Opecoelidae) from the Ouchita madtom, Noturus lachneri, and the

banded sculpin, Cottus carolinae, from Arkansas. J. Parasitol. 104, 145–156.

Fessler, J.L., Westneat, M.W., 2007. Molecular phylogenetics of the butterflyfishes

(Chaetodontidae): taxonomy and biogeography of a global coral reef fish family. Mol.

Phylogenet. Evol. 45, 50–68.

Gochfeld, D.J., 2004. Predation-induced morphological and behavioral defenses in a hard coral:

implications for foraging behaviour of coral-feeding butterflyfishes. Mar. Ecol. Prog. Ser.

267, 145–158.

Harmelin-Vivien, M.L., Bouchon-Navaro, Y., 1983. Feeding, diets and significance of coral feeding

among chaetodontid fishes in Moorea (French Polynesia). Coral Reefs 2, 119–127.

Hendrix, S.S., 1973. The life history and biology of Plagioporus hypentelii Hendrix, 1973

(Trematoda: Opecoelidae). J. Parasitol. 64, 606–612.

Hildebrand, J., Sitko, J., Zaleśny, G., Jeżewski, W., Laskowski, Z., 2016. Molecular characteristics

of representatives of the genus Brachylecithum Shtrom, 1940 (Digenea, Dicrocoeliidae) with

comments on life cycle and host specificity. Parasitol. Res. 115, 1417–1425.

Huston, D.C., Cutmore, S.C., Cribb, T.H., 2016. The life-cycle of Gorgocephalus yaaji Bray &

Cribb, 2005 (Digenea: Gorgocephalidae) with a review of the first intermediate hosts for the

superfamily Lepocreadioidea Odhner, 1905. Syst. Parasitol. 93, 653–665.

Jousson, O., Bartoli, P., Pawlowski, J., 1999. Molecular identification of developmental stages in

Opecoelidae (Digenea). Int. J. Parasitol. 29, 1853–1858.

Jousson, O., Bartoli, P., 2000. The life cycle of Opecoeloides columbellae (Pagenstecher, 1863) n.

comb. (Digenea: Opecoelidae): evidence from molecules and morphology. Int. J. Parasitol.

30, 747–760.

Keller, A., Schleicher, T., Schultz, J., Muller, T., Dandekar, T., Wolf, M., 2009. 5.8S–28S rRNA

interaction and HMM-based ITS2 annotation. Gene 430, 50–57.

Lanfear, R., Calcott, B., Ho, S.Y.W., Guindon, S., 2012. PartitionFinder: combined selection of

partitioning schemes and substitution models for phylogenetic analyses. Mol. Biol. and

Evol. 29, 1695–1701.

Leiva, N.V., López, Z., González, M.T., Muñoz, G., 2017. Determining intermediate hosts for

Opecoelidae and Microphallidae species (Platyhelminthes: Trematoda) in the Southeastern

Pacific Coast, using molecular markers. J. Parasitol. 103, 132–137.

Littlewood, D.T.J., McDonald, S.M., Gill, A.C., Cribb, T.H., 2004. Molecular phylogenetics of

Chaetodon and the Chaetodontidae (Teleostei: Perciformes) with reference to morphology.

Zootaxa 779, 1–20.

Littlewood, D.T.J., Bray, R.A., Waeschenbach, A., 2015. Phylogenetic patterns of diversity in

cestodes and trematodes. In: Morand, S., Krasnov, B., Littlewood, D.T.J. (Eds). Parasite

Diversity and Diversification: Evolutionary Ecology Meets Phylogenetics. Cambridge

University Press, Cambridge, UK, pp. 304–319.

Lucas, T., O’Brien, E.K., Cribb, T.H., Degnan, B.M., 2005. Digenean trematodes infecting the

tropical abalone Haliotis asinina have species-specific cercarial emergence patterns that

follow daily or semilunar spawning cycles. Mar. Bio. 148, 285–292.

Martin, S.B., Cutmore, S.C., Cribb, T.H., 2017. Revision of Neolebouria Gibson, 1976 (Digenea:

Opecoelidae), with Trilobovarium n. g., for species infecting tropical and subtropical

shallow-water fishes. Syst. Parasitol. 94, 307–338.

Martin, S.B., Cribb, T.H., Cutmore, S.C., Huston, D.H., 2018a. The phylogenetic position of

Choerodonicola Cribb, 2005 (Digenea: Opecoelidae) with a partial life cycle for a new

species from the blue-barred parrotfish Scarus ghobban Forsskål (Scaridae) in Moreton Bay,

Australia. Syst. Parasitol. 95, 337–352.

Martin, S.B., Crouch, K., Cutmore, S.C., Cribb, T.H., 2018b. Expansion of the concept of the

Opistholebetinae Fukui, 1929 (Digenea: Opecoelidae Ozaki, 1925), with Magnaosimum

brooksae n. g., n. sp. from Tripodichthys angustifrons (Hollard) (Tetradontiformes:

Triacanthidae) in Moreton Bay, Australia. Syst. Parasitol. 95, 121–132.

Martin, S.B., Cutmore, S.C., Cribb, T.H., 2018c. Revision of Podocotyloides Yamaguti, 1934

(Digenea: Opecoelidae), resurrection of Pedunculacetabulum Yamaguti, 1934 and the

naming of a cryptic opecoelid species. Syst. Parasitol. 95, 1–31.

McCoy, O.R, 1929. The life-history of a marine trematode, Hamacreadium mutabile Linton, 1910.

Parasitol. 21, 220–225.

McCoy, O.R, 1930. Experimental studies on two fish trematodes of the genus Hamacreadium

(Family Allocreadiidae). J. Parasitol. 17, 1–3.

McNamara, M.K.A., Adlard, R.D., Bray, R.A., Sasal, P., Cribb, T.H., 2012. Monorchiids

(Platyhelminthes: Digenea) of chaetodontid fishes (Perciformes): biogeographical patterns

in the tropical Indo-West Pacifc. Parasitol. Int. 61, 288–306.

Meenakshi, M., Madhavi, R., Swarnakumari, V.G.M., 1993. The life-cycle of Helicometra gibsoni

n. sp. (Digenea: Opecoelidae). Syst. Parasitol. 25, 63–72.

Nagelkerken, I., van der Velde, G., Wartenbergh, S L.J., Nugues, M.M., Pratchett, M.S., 2009.

Cryptic dietary components reduce dietary overlap among sympatric butterflyfishes

(Chaetodontidae). J. Fish Biol. 75, 1123–1143.

Olson, P.D., Cribb, T.H., Tkach, V.V., Bray, R.A., Littlewood, D.T.J., 2003. Phylogeny and

classification of the Digenea (Platyhelminthes: Trematoda). Int. J. Parasitol. 33, 733–755.

Pratchett, M.S., 2005. Dietary overlap among coral-feeding butterflyfishes (Chaetodontidae) at

Lizard Island, north Great Barrier Reef. Mar. Biol. 148, 373–382.

Pratchett, M.S., 2014. Feeding preferences and dietary specialisation among obligate coral-feeding

butterflyfishes. In: Pratchett, M.S., Berumen, M.L., Kapoor, B.G. (Eds), Biology of

Butterflyfishes. CRC Press: Boca Raton, Florida, USA, pp. 140–179.

Pritchard, M.H., 1966. Studies of digenetic trematodes of Hawaiian fishes: Family Opecoelidae

Ozaki, 1925. Zool. Jahrb. Abt. Syst. Ökol. Geogr. Tiere 93, 173–202.

Powell, A., Jones, T., Smith, D.J., Jompa, J., Bell, J.J., 2015. Spongivory in the Wakatobi National

Park, Southeast Sulawesi, Indonesia. Pac. Sci. 69, 487–508.

Reavis, R.H., Copus, J.M., 2011. Monogamy in a feeding generalist, Chaetodon trichrous, the

endemic Tahitian butterflyfish. Environ. Biol. Fish. 92, 167–179.

Rigby, M.C., Lo, C.M., Cribb, T.H., Euzet, L., Faliex, E., Galzin, R., Holmes, J.C., Morand, M.,

1999. Checklist of the parasites of coral reef fishes from French Polynesia, with

considerations on their potential role in these fish communities. Cybium 23, 273–284.

Rima, M., Marzoug, D., Pérez-del-Olmo, A., Kostadinova, A., Bouderbala, M., Georgieva, S.,

2017. New molecular and morphological data for opecoelid digeneans in two Mediterranean

sparid fishes with descriptions of Macvicaria gibsoni n. sp. and M. crassigula (Linton,

1910) (sensu stricto). Syst. Parasitol. 94, 739–763.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D., Darling, A., Höhna, S., Larget, B., Liu, L.,

Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efficient Bayesian phylogenetic

inference and model choice across a large model space. Syst. Biol. 61, 539–542.

Sano, M., 1989. Feeding habits of Japanese butterflyfishes (Chaetodontidae). Environ. Biol. Fish.

25, 195–203.

Schell, S.C., 1975. The life history of Plagioporus shawi (McIntosh, 1939) (Trematoda:

Opecoelidae), an intestinal parasite of salmonid fishes. J. Parasitol. 61, 899–905.

Schell, S.C., 1976. The life history of Nezpercella lewisi Schell, 1974 (Trematoda: Opecoelidae), a

parasite of the northern squawfish and the smallmouth bass. J. Parasitol. 62, 894–898.

Shedko, M. B., Sokolov, S. G., Atopkin, D. M., 2015. The first record of Dimerosaccus

oncorhynchi (Trematoda: Opecoelidae) in fishes from rivers of Primorsky Territory, Russia,

with a discussion on its taxonomic position using morphological and molecular data.

Parazitologiya 49, 171–189.

Shoop, W.L., 1989. Systematic analysis of the Diplostomidae and Strigeidae (Trematoda). J.

Parasitol. 79, 21–32.

Sinitsin, D. 1931. Studien über die phylogenie der trematoden. IV. The life histories of Plagioporus

siliculus and Plagioporus virens, with special reference to the origin of the Digenea. Zeit.

wiss. Zool. 138, 409–456.

Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large

phylogenies. Bioinformatics 30, 1312–1313.

Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular

evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725–2729.

Thompson, A.B., Margolis, L., 1987. Descriptions of Neolebouria tinkerbellae n. sp. (Trematoda:

Digenea: Opecoelidae) from experimental fish hosts, and of metacercariae of N. tinkerbellae

and an unidentified digenean from Pandalus jordani (Decapoda: Penaeidae) from the

Pacific coast of Canada. Can. J. of Zool., 65, 188–193.

Tkach, V.V., Kudlai, O., Kostadinova, A., 2016. Molecular phylogeny and systematics of the

Echinostomatoidea Looss, 1899 (Platyhelminthes: Digenea). Int. J. Parasitol. 46, 171–185.

Tkach, V.V., Pawlowski, J., Mariaux, J., Swiderski, Z., 2001. Molecular phylogeny of the suborder

Plagiorchiata and its position in the system of Digenea. In: Littlewood, D.T.J., Bray, R.A.

(Eds), Interrelationships of the Platyhelminthes.. Taylor and Francis, London, pp. 186–193.

Toman, G., 1992. Digenetic trematodes of marine teleost fishes from the Seychelles, Indian Ocean.

III. Acta Parasitol. 37, 119–126.

Vera, M.,Banks, S., 2009. Health status of the coral communities of the northern Galapágos Islands

Darwin, Wolf and Marchena. Galapágos Res. 66, 65–74.

Willis, B.L., Page, C.A., Dinsdale, E.A., 2004. Coral disease on the Great Barrier Reef. In:

Rosenburg, E., Loya, Y. (Eds), Coral Health and Disease. Springer-Verlag, Germany, pp.

69–104.

Work, T.M., Aeby, G.S., Lasne, G., Tribollet, A., 2014. Gross and microscopic pathology of hard

and soft corals in New Caledonia. J. Invert. Path. 120, 50–58.

Yano, A., Urabe, M., 2017. Larval stages of Neoplagioporus elongatus (Goto and Ozaki, 1930)

(Opecoelidae: Plagioporinae), with notes on potential second intermediate hosts. Parasitol.

Int. 66, 181–185.

Yoshida, R., Urabe, M., 2005. Life cycle of Coitocoecum plagiorchis (Trematoda: Digenea:

Opecoelidae). Parasitol. Int. 54, 237–242.

Yamaguti, S. (1970). Digenetic trematodes of Hawaiian fishes. Interscience, New York.

Yamashiro, H. (2004). Coral Disease. In: Coral Reefs of Japan. Minister for the Environment,

Japanese Coral Reef Society, Japan, pp. 56–60.

Figure captions

Fig. 1. Map of French Polynesia. Chaetodontids were collected from off labelled islands.

Fig. 2. Lateral perspectives of whole-mount (A) and terminal genitalia (B) from voucher specimens of Polypipapiliotrema stenometra (Pritchard, 1966) n. comb.ex Chaetodon quadrimaculatus off

O‘ahu, Hawaii, and of holotype (C) and terminal genitalia of paratype (D) for Polypipapiliotrema heniochi n. sp. ex Heniochus chrysostomus off Mo'orea, Society Islands. Scale-bars: A and C, 500

µm; B and D, 300 µm.

Fig. 3. Morphometric comparison of Polypipapiliotrema species. Boxes represent Q1 to Q3 with the mean indicated by a filled diamond; ‘whiskers’ (error bars) indicate 1.5 times the inter-quartile range, beyond which outliers are depicted as open circles. Ps, P. stenometra; Phe, P. heniochi; Pha,

P. hadrometra; Po, P. ovatheculum; Pc, P. citerovarium.

Fig. 4. Lateral perspectives of whole-mount (A) and terminal genitalia (B) from holotype of

Polypipapiliotrema hadrometra n. sp. ex Chaetodon lunula off Maria, Austral Islands, and of whole-mount (A) and terminal genitalia (B) from holotype of Polypipapiliotrema ovatheculum n. sp. ex Chaetodon flavirostris off Ra'ivāvae, Austral Islands. Scale-bars: A and C, 500 µm; B and D,

300 µm.

Fig. 5. Lateral perspectives of whole-mount (A) and terminal genitalia (B) from holotypes of

Polypipapiliotrema citerovarium n. sp. ex Chaetodon quadrimaculatus off Ra'ivāvae, Austral

Islands, and of terminal genitalia (C) and whole-mount (D) from voucher specimens ex C. quadrimaculatus off the Mangareva island group, Gambier Islands. Scale-bars: A and D, 500 µm; B and C, 100 µm.

Fig. 6. Neighbour-joining analysis of the ITS2 rDNA alignment for species of Polypipapiliotrema

Martin, Cutmore & Cribb n. gen. including all replicate sequences generated and available.

GenBank sequences DQ083434 and MF926404 were originally identified as Podocotyloides stenometra by Lucas et al. (2003) and Martin et al. (2018), respectively. The scale bar indicates uncorrected base-pair differences.

Fig. 7. Morphometric comparison of Polypipapiliotrema citerovarium specimens collected from chaetodontids off the Mangareva island group in the Gambier Islands and off mainly Raivavae

(some specimens from fishes off Maria) in the Austral Islands. Boxes represent first and third quartiles with the mean indicated by a filled diamond; ‘whiskers’ (error bars) indicate 1.5 times the inter-quartile range, beyond which outliers are depicted as open circles.

Fig. 8. Bayesian inference analysis of concatenated 28S + 18S rDNA alignment for the

Opecoelidae. Support values less than 65% are not shown. The scale bar indicates the expected number of substitutions per site.

Table 1. GenBank accession numbers for the sequence data used in phylogenetic analyses. Taxon GenBank ID GenBank ID Reference (18S) (28S) Allopodocotyle epinepheli (Yamaguti, 1942) KU320585 KU320598 Bray et al. (2016) Allopodocotyle margolisi Gibson, 1955 KU320583 KU320596 Bray et al. (2016) Allopodocotyle sp. A KU320586 KU320599 Bray et al. (2016) Allopodocotyle sp. B KU320594 KU320607 Bray et al. (2016) Anomalotrema koiae Gibson & Bray, 1984 KU320582 KU320595 Bray et al. (2016) Bathycreadium brayi Pérez-del-Olmo, Dallarés, Carrassón & Kostadinova, 2014a JN085948 Constenla et al. (2011) Bentholebouria blatta Bray & Justine, 2009 KU320593 KU320606 Bray et al. (2016) Bentholebouria colubrosa Andres, Pulis & Overstreet, 2014 KJ001207 Andres et al. (2014a) Buticulotrema thermichthysi Bray, Waeschenbach, Dyal, Littlewood & Morand, 2014 KF733984 KF733987 Bray et al. (2012) Cainocreadium labracis (Dujardin, 1845) JQ694144 Born-Torrijos et al. (2012) Cainocreadium lintoni (Siddiqi & Cable, 1960) KJ001208 Andres et al. (2014a) Choerodonicola arothokoros Martin, Cribb, Cutmore & Huston, 2018 MG844417 MG844418 Martin et al. (2018a) Choerodonicola renko Machida, 2014 MG844420 MG844421 Martin et al. (2018a) Dimerosaccus oncorhynchi (Eguchi, 1931) FR870262 Shedko et al. (2015) Gaevskajatrema halosauropsi Bray & Campbell, 1996 AJ287514 AY222207 Cribb et al. (2001); Olson et al. (2003) Gaevskajatrema perezi (Mathias, 1926) AF184255 Tkach et al. (2001) Hamacreadium cribbi Bray & Justine, 2016b KU320590 KU320603 Bray et al. (2016) Hamacreadium mutabile Linton, 1910 KJ001209 Andres et al. (2014a) Hamacreadium sp.c KU320588 KU320601 Bray et al. (2016) Helicometra epinepheli Yamaguti, 1934d KU320584 KU320597 Bray et al. (2016) Helicometra equilata (Manter, 1933)e KU320587 KU320600 Bray et al. (2016) Helicometra manteri Andres, Ray, Pulis, Curran & Overstreet, 2014 KJ701238 Andres et al. (2014b) Maculifer sp. AY222109 AY222211 Olson et al. (2003) Macvicaria bartolii Antar, Georgieva, Gargouri & Kostadinova, 2014 KR149471 Antar et al. (2015) Macvicaria crassigula (Linton, 1910) KJ701237 Andres et al. (2014b) Macvicaria dubia (Stossich, 1905) KR149470 Antar et al. (2015) Macvicaria gibsoni Rima, Marzoug, Pérez-del-Olmo, Kostadinova, Bouderbala & MF166845 Rima et al. (2017) Georgieva, 2017 Macvicaria maamouriae Antar, Georgieva, Gargouri & Kostadinova, 2014 KR149468 Antar et al. (2015) Macvicaria macassarensis (Yamaguti, 1952) AJ287533 AY222208 Cribb et al. (2001); Olson et al. (2003) Macvicaria magellanica Laskowski, Jeżewski & Zdzitowiecki, 2013 KU212191 Hildebrand et al. (2016) Macvicaria mormyri (Stossich, 1885) AF184256 Tkach et al. (2001) Macvicaria obovata (Mollin, 1859) JQ694146 Born-Torrijos et al. (2012) Magnaosimum brooksae Martin, Krouch, Cutmore & Cribb, 2018 MG813906 MG813907 Martin et al. (2018b) Neolebouria lanceolata (Price, 1934) KJ001210 Andres et al. (2014a) Neoplagioporus ayu (Takahashi, 1928) KX553947 Fayton and Andres (2016) Neoplagioporus elongatus (Goto & Ozaki, 1930) KX553948 Fayton and Andres (2016) Neoplagioporus zacconis (Yamaguti, 1934) KX553949 Fayton and Andres (2016) Opistholebes amplicoelus Nicoll, 1915 AJ287550 AY222210 Cribb et al. (2001); Olson et al. (2003) Opecoeloides furcatus (Bremser in Rudolphi, 1819) AF151937 Tkach et al. (2001) Opecoeloides fimbriatus (Linton, 1934) KJ001211 Andres et al. (2014a) Pacificreadium serrani (Nagaty & Abdel-Aal, 1962) KU320589 KU320602 Bray et al. (2016) Pedunculacetabulum inopinipugnus Martin, Cutmore & Cribb, 2017 MF805699 MF805700 Martin et al. (2018c) Peracreadium idoneum (Nicoll, 1909) AJ287558 AY222209 Cribb et al. (2001); Olson et al. (2003) Plagioporus aliffi Fayton, Choudhury, McAllister & Robison, 2017 KX905056 Fayton et al. (2017) Plagioporus boleosomi (Pearse, 1924) KX553953 Fayton and Andres (2016) Plagioporus carolini Fayton, McAllister, Robison & Cannior, 2018 MG214680 Fayton et al. (2018) Plagioporus chiliticorum (Barger & Esch, 1999) KX553943 Fayton and Andres (2016) Plagioporus fonti Fayton, Choudhury, McAllister & Robison, 2017 KX905054 Fayton et al. (2017) Plagioporus hageli Fayton & Andres, 2016 KX553950 Fayton and Andres (2016) Plagioporus ictaluri Fayton, McAllister, Robison & Cannior, 2018 MG214679 Fayton et al. (2018) Plagioporus kolipinskii Tracey, Choudhury, Cheng & Ghosh, 2009 KX553952 Fayton and Andres (2016) Plagioporus limus Fayton, Choudhury, McAllister & Robison, 2017 KX905055 Fayton et al. (2017) Plagioporus loboides (Curran, Overstreet & Tkach, 2007) f EF523477 Curran et al. (2007) Plagioporus shawi (McIntosh, 1939) KX553951 Fayton and Andres (2016) Plagioporus sinitsini Mueller, 1934 KX553944 Fayton and Andres (2016) Podocotyloides australis Martin, Cutmore & Cribb, 2017 MF805695 MF805696 Martin et al. (2018c) Podocotyloides brevis Andres & Overstreet, 2013 KJ001212 Andres et al. (2014a) Podocotyloides gracilis (Yamaguti, 1952) Pritchard, 1966 MF805692 MF805693 Martin et al. (2018c) Podocotyloides parupenei (Manter, 1962) Pritchard, 1966 MF926408 MF926409 Martin et al. (2018c) Polypipapiliotrema citerovarium sp. nov. MH823962/3 MH823957/8 present study Polypipapiliotrema hadrometra sp. nov. MH823960 MH823955 present study Polypipapiliotrema heniochi sp. nov.g MF926405 MF926406 Martin et al. (2018c) Polypipapiliotrema ovatheculum sp. nov. MH823961 MH823956 present study Polypipapiliotrema stenometra (Pritchard, 1966) comb. nov. MH823959 MH823954 present study Propycnadenoides philippinensis Fischthal & Kuntz, 1964 KU320591 KU320604 Bray et al. (2016) Pseudopecoeloides tenuis Yamaguti, 1940 KU320592 KU320605 Bray et al. (2016) Pseudopycnadena fischthali Saad-Fares & Maillard, 1986 MF166851 Rima et al. (2017) Pseudopycnadena tendu Bray & Justine, 2007 FJ788506 Bray and Justine (2009) Trilobovarum parvvatis Martin, Cutmore & Cribb, 2017 KY551561 KY551562 Martin et al. (2017) Urorchis acheilognathi Yamaguti, 1934 KX553945 Fayton and Andres (2016) Urorchis goro Ozaki, 1927 KX553946 Fayton and Andres (2016) Outgroup Gorgocephalus yaaji Bray & Cribb, 2005 KU951489 Huston et al. (2016) Preptetos caballeroi Pritchard, 1960 AJ287563 AY222236 Cribb et al. (2001); Olson et al. (2003) Stephanostomum pristis (Deslongchamps, 1824) DQ248209 DQ248222 Bray et al. (2005) Zalophotrema hepaticum Stunkard & Alvey, 1929 AJ224884 AY222255 Cribb et al. (2001); Olson et al. (2003) aRegistered as Bathycreadium elongatum (Maillard, 1970) Bray, 1973, see Pérez-del-Olmo et al. (2014). bRegistered as Hamacreadium sp., see Bray and Justine (2016). cRegistered as Hamacreadium mutabile, see Bray and Justine (2016). dRegistered as Helicometra fasciata (Rudolphi, 1819) Odhner, 1902. eRegistered under the junior synonym Helicometra boseli Nagaty, 1956, see Blend and Dronen (2015a). fRegistered under the junior synonym Plagiocirrus loboides Curran, Overstreet & Tkach, 2007, see Fayton and Andres (2016). gRegistered under the junior synonym Podocotyloides stenometra Pritchard, 1966.

Table 2. Sampling effort of chaetodontids by region. Detected infections of species of Polypipapiliotrema are highlighted, with species listed. Chaetodontid diets based on Harmelin-Vivien and Bouchon-Navaro (1983), Bouchon-Navaro (1986), Sano (1989), Cox (1994), Pratchett (2005, 2011), Cole et al. (2008) and Reavis and Copus (2011). Phylogenetic clades for species of Chaetodon are as per Littlewood et al. (2004), Fessler and Westneat (2007) and Bellwood et al. (2010).

Clade Dieta Hw Aus Gam Mq Soc Tua Total Infectionsb Chaetodon citrinellus 2 F - 8 - 8 66 15 97 Pc fremblii 2 N 1 - - - - - 1 kleinii 2 F 4 - - - - - 4 mertensii 2 F - 3 3 1 - - 7 Pc miliaris 2 N 1 - - - - - 1 multicinctus 2 O 16 - - - - - 16 pelewensis 2 O - 10 6 5 19 9 19 Pc quadrimaculatus 2 F 6 7 3 3 32 6 57 Ps, Pc trichrous 2 F - - - 10 2 - 12 unimaculatus 2 F - 7 4 1 20 1 33 bennetti 3 O - 4 - - 1 2 7 lunulatus 3 O - 4 5 - 29 10 48 ornatissimus 3 O - 8 1 2 31 5 47 reticulatus 3 O - 9 3 2 22 11 47 Pc trifascialis 3 O - 4 4 - 3 6 17 auriga 4 N - 8 10 3 16 12 49 Po ephippium 4 N - 3 1 1 15 10 30 flavirostris 4 F - 10 4 - 1 - 15 Po lineolatus 4 N - 1 1 lunula 4 F - 3 2 6 15 10 36 Pha ulietensis 4 F - 2 2 - 11 9 24 vagagundus 4 N - 3 - - 35 - 38 declivis ? ? - - - 2 - - 2 Forcipiger flavissimus N - 16 5 10 36 8 75 longirostris N - 6 - 1 - - 7 Hemitaurichthys polylepis N - 6 5 - - 3 14 thompsoni N - - - 3 - - 3 Heniochus chrysostomus F - 3 4 - 29 11 47 Phe monoceros N - 5 1 - 1 5 12 Phe

Hw, Hawai‘i; Aus, Austral Islands; Gam, Mangareva group in the Gambier Islands; Mq, Marquesas Islands; Soc, Moorea in the Society Islands; Tua, Tuamotu Islands. aO, obligate corallivore, corals constitute >80% of diet; F, facultative corallivore, generalists preying significantly on corals; N, non-corallivore, corals constitute <10% of diet. bPc, P. citerovarium; Pha, P. hadrometra; Phe, P. heniochi; Po, P. ovatheculum; Ps, P. stenometra.

Table 3. Morphometric data for species of Polypipapiliotrema, expressed in micrometres, as percentages or as ratios. Body depth is measured at the level of the ovary. Cirrus-sac length is provided as an estimate of actual length, but also as the occupied proportion of body length. Egg measurements represent the average of multiple subsamples per specimen.

P. stenometra P. heniochi P. hadrometra P. ovatheculum P. citerovarium n 8 15 7 2 27 BL 978–1,594 (1,320) 1,688–2,878 (2,156) 2,239–4,938 (3,026) 2,016, 2,484 (2,250) 888–1,714 (1,352) BD 681–1,029 (830) 152–243 (196) 214–325 (259) 223, 235 (229) 125–203 (163) FBL 150–239 (183) 211–364 (261) 315–558 (421) 338, 367 (353) 158–308 (235) FBL/BL 11–18 (14)% 10–14 (12)% 11–17 (14)% 15, 17 (16)% 13–21 (17)% OSL 55–79 (65) 66–114 (82) 105–195 (149) 118, 152 (135 ) 62–101 (83) OSD 60–83 (72) 78–113 (91) 109–178 (137) 123, 132 (128) 55–112 (94) OSD/OSL 1–1.22 (1.11) 0.98–1.22 (1.12) 0.85–1.07 (0.94) 0.87, 1.04 (0.96) 0.87–1.32 (1.11) VSL 90–156 (132) 120–272 (157) 198–299 (261) 220, 224 (222) 106–208 (153) VSD 80–121 (95) 88–189 (122) 183–306 (228) 175, 231 (203) 99–172 (113) VSL/VSD 0.74–1.67 (1.4) 1.06–1.72 (1.29) 0.96–1.32 (1.2) 0.97, 1.26 (1.11) 0.89–1.43 (1.2) VSL/OSL 1.55–2.34 (2.02) 1.7–2.39 (1.89) 1.51–2 (1.78) 1.47, 1.86 (1.67) 1.32–2.38 (1.85) VSD/OSD 1.08–1.75 (1.35) 1.09–1.67 (1.33) 1.44–1.72 (1.63) 1.42, 1.75 (1.57) 1.25–1.8 (1.42) Prepharynx 9–19 (15) 22–38 (31) 10–30 (20) 14, 25 (20) 12–38 (23) PhL 64–89 (76) 62–115 (77) 90–156 (121) 104, 115 (110) 73–116 (94) PhD 57–77 (68) 69–106 (85) 97–149 (118) 87, 97 (92) 73–104 (86) PhL/PhD 0.93–1.24 (1.11) 0.8–1.14 (0.9) 0.93–1.2 (1.03) 1.19, 1.2 (1.19) 0.94–1.26 (1.11) OSL/PhL 0.73–0.95 (0.87) 0.96–1.17 (1.08) 1.06–1.34 (1.22) 1.13, 1.32 (1.23) 0.74–1.1 (0.88) OSD/PhD 0.98–1.12 (1.06) 0.98–1.26 (1.08) 1.1–1.37 (1.16) 1.36, 1.41 (1.39) 0.71–1.27 (1.09) Oesophagus/BL 5–8 (6)% 5–8 (6)% 3–6 (4)% 4, 5 (4)% 4–8 (6)% PreBif/BL 14–22 (17)% 12–17 (14)% 10–16 (14)% 15, 16 (15)% 16–25 (19)% PreVit/BL 26–47 (33)% 12–27 (22)% 29–37 (33)% 32, 37 (34)% 25–38 (31)% PostC/BL 2–5 (3)% 4–6 (5)% 4–6 (5)% 4, 6 (5)% 4–9 (6)% ATL 45–97 (70) 101–219 (146) 118–259 (175) 143, 150 (147) 61–131 (102) ATD 29–59 (49) 75–157 (104) 105–187 (134) 81, 124 (103) 45–119 (89) ATL/ATD 1.15–1.8 (1.41) 1.25–1.66 (1.4) 1.08–1.61 (1.3) 1.2, 1.77 (1.49) 0.92–1.58 (1.18) PTL 42–112 (76) 112–246 (161) 127–283 (191) 158, 175 (167) 72–157 (102) PTD 28–60 (50) 80–166 (109) 103–172 (133) 110, 118 (114) 45–117 (88) PTL/PTD 1.16–1.93 (1.5) 1.26–1.88 (1.48) 1.08–1.72 (1.43) 1.44, 1.48 (1.46) 1.03–1.82 (1.33) PTL/ATL 0.93–1.2 (1.08) 1.05–1.18 (1.11) 0.93–1.28 (1.1) 1.10, 1.17 (1.14) 0.88–1.29 (1.12) PTD/ATD 0.97–1.07 (1.02) 0.91–1.15 (1.05) 0.91–1.12 (1) 0.45, 1.36 (1.15) 0.74–1.22 (0.99) PreT 62–78 (69)% 56–69 (63)% 64–73 (62)% 71, 73 (72)% 57–70 (64)% PostT 12–24 (18)% 18–23 (21)% 15–21 (18)% 21% 15–27 (21)% CSL 302–477 (400) 323–582 (425) 553–1,206 (785) 652, 742 (697) 184–379 (297) CSD 28–43 (35) 21–58 (35) 44–97 (61) 52, 62 (57) 21–45 (34) CSL/BL 22–35 (28)% 18–21 (20)% 22–27 (24)% 28, 31 (29)% 19–24 (21)% SVL/CSL 25–54 (36)% 31–56 (46)% 21–39 (33)% 37, 38 (38)% 25–59 (38)% PPL/CSL 14–24 (17)% 8–18 (12)% 13–29 (21)% 20, 22 (21)% 7–20 (11)% EDL/CSL 33–59 (46)% 30–60 (40)% 37–52 (44)% 42, 43 (42)% 26–61 (48)% Genital atrium 13–45 (29) 15–37 (25) 51–89 (64) 27, 38 (33) 13–38 (22) PreGP 8–11 (10)% 6–9 (8)% 7–10 (9)% 9, 10 (9)% 10–16 (12)% OvL 45–62 (53) 63–123 (87) 77–182 (123) 114, 139 (127) 43–90 (67) OvD 35–46 (40) 64–108 (82) 79–167 (113) 90, 110 (100 ) 40–92 (70) PreOv/BL 57–72 (63)% 50–61 (56)% 60–65 (62)% 60, 64 (62)% 52–67 (59)% Ov to VS 31–49 (41)% 32–44 (39)% 26–44 (38)% 39, 43 (41)% 19–42 (31)% Ov to CS 16–29 (23)% 23–31 (28)% 23–29 (27)% 21, 22 (22)% 18–33 (25)% Egg count 4–17 (10) 30–170 (100) 40–160 (85) 56, 64 (60) 4–45 (20) Egg length 50–55 (52) 50–58 (54) 54–60 (57) 60, 63 (62) 48–59 (55) Egg width 30–36 (32) 34–39 (36) 35–37 ( 36) 35, 36 (35) 34–40 (37)

L, length; D, depth; B, body; FB, forebody; OS, oral sucker; VS, ventral sucker; Ph, pharynx; Bif, bifurcal zone; Vit, vitelline zone; C, caecal zone; AT, anterior testis; PT, posterior testis; CS, cirrus-sac; SV, seminal vesicle; PP, pars-prostatica; ED, ejaculatory duct; GP, genital pore; Ov, ovary.

Table 4. Molecular differences (number of base-pairs) in the ITS2 (429 bases), 28S (1,307 bases) and 18S (1,801) rDNA marker regions among seven uncovered genotypes of Polypipapiliotrema n. gen. species: 1, Polypipapiliotrema stenomerta; 2, Polypipapiliotrema heniochi n. sp.; 3, Polypipapiliotrema hadrometra; 4, Polypipapiliotrema ovatheculum n. sp.; 5, Polypipapiliotrema citerovarium n. sp. Raivavae /Maria variant; 6, Polypipapiliotrema citerovarium n. sp. Tubuai variant; and 7, Polypipapiliotrema citerovarium n. sp. Gambier Islands variant (ITS2 data only).

2 3 4 5 6 7

ITS2 28S 18S ITS2 28S 18S ITS2 28S 18S ITS2 28S 18S ITS2 28S 18S ITS2 28S 18S 1 5 8 4 4 9 4 6 11 2 8 14 5 9 13 5 9 na na 2 - - - 9 9 4 11 10 2 10 15 7 11 14 7 11 na na 3 ------9 11 2 12 16 7 11 15 7 13 na na 4 ------8 11 5 9 9 5 9 na na 5 ------1 3 0 1 na na 6 ------2 na na

Graphical abstract

Highlights

 Diversification of the Opecoelidae appears driven by intermediate host switches

 A new genus and subfamily are suggested for the only lineage of trematodes known to infect

corals

 Trematodiasis in corals is likely caused by a complex of many species

 Four new opecoelid species from butterflyfishes in French Polynesia are described