Integrated analyses

For the Pomacanthidae, 37 individual fishes were examined, representing 9 species; specimens relevant to this study were found in 6 of the 9 species, from Heron Island, Lizard Island and Yonge Reef on the Great Barrier Reef, Queensland, and from Ningaloo Reef, Western Australia. For the Ostraciidae, 137 individual fishes were examined, representing 5 species; specimens relevant to this study were found in 2 of the 5 species, from Heron Island, Lizard Island, Ningaloo Reef, and Okinawa. Specimens collected from these two fish families were broadly consistent with concepts of three currently recognised genera: Antorchis Linton, 1911, Paradiscogaster Yamaguti, 1934, and Parayamagutia Machida, 1971.

Specimens consistent with the current generic diagnosis of Antorchis were recovered from four pomacanthid species: the emperor angelfish, Pomacanthus imperator (Bloch); the blue angelfish, Pomacanthus semicirculatus (Cuvier); the sixbar angelfish, Pomacanthus sexstriatus (Cuvier); and the yellowface angelfish, Pomacanthus xanthometopon (Bleeker). cox1 sequences were generated from specimens representative of all host-parasite-locality combinations of the new samples, and also from a single specimen of Antorchis nasonis Cribb, Bray, Hall & Cutmore, 2015 from among the material reported by Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) and from four specimens of Antorchis pomacanthi (Hafeezullah and Siddiqi, 1970) Machida, 1975 from New Caledonia reported by Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015). Phylogenetic analysis of these data (Figure 1) results in three principal clades that differ at 7–102 bp (p-distance = 18.35–21.52%). The first principal clade is represented by specimens from three species of Pomacanthus (Lacépède) from the Great Barrier Reef, Ningaloo Reef and New Caledonia; this clade is identified as representing Antorchis pomacanthi, matching published genetic data for this species for both the ITS2 and 28S regions. Sequences of Antorchis pomacanthi form four clades in the cox1 dataset; significant variation, interpreted as intraspecific, was found between specimens from Heron Island + Lizard Island, Ningaloo Reef (two clades) and New Caledonia, with the four clades differing at 25–33 bp (p-distance = 5.27–6.96%). The second principal clade is represented by just the single sequence of Antorchis nasonis. The remaining principal clade is represented by specimens from two species of Pomacanthus from the northern Great Barrier Reef (Lizard Island and Yonge Reef). Specimens of this final clade form two secondary clades that differ at 35–36 bp (p-distance = 7.38–7.6%), which we interpret as two new species based on combined morphological and molecular evidence.

Figure 1.

Phylogram from the unrooted neighbour-joining analysis of the cox1 mtDNA dataset for species that agree with the current generic diagnosis of Antorchis Linton, 1911. Bootstrap support values are shown at the nodes, with values of <85 not shown. The scale bar indicates the number of base differences.

Figure 1. Long description

A phylogram with a root on the left branching into two primary lineages.

Top Lineage:

This large clade is dominated by Neoparantorchis pomacanthi. It splits into two main sub-clades.

- The upper sub-clade, supported by a bootstrap value of 99, contains numerous specimens from Lizard Island, Heron Island, and various Pomacanthus hosts.

- The lower sub-clade, also supported by a bootstrap value of 99, contains specimens from New Caledonia (bootstrap 91 and 88) and Ningaloo Reef.

Bottom Lineage:

This lineage branches into three distinct species groups.

- The first branch leads to Nasoantorchis nasonis from Naso annulatus at Heron Island.

- The second branch, supported by a bootstrap value of 99, leads to Parantorchis parviovatus from Pomacanthus imperator at Yonge Reef.

- The third branch, supported by a bootstrap value of 99, contains multiple specimens of Parantorchis vaili from Pomacanthus sexstriatus and Pomacanthus imperator at Lizard Island and Yonge Reef.

A scale bar at the bottom left indicates 10 base differences. Bootstrap support values of 99 are frequently noted at major nodes.

Specimens consistent with the current generic diagnosis of Paradiscogaster were recovered from five pomacanthid species and two ostraciid species. For the Pomacanthidae, specimens were found in the keyhole angelfish, Centropyge tibicen (Cuvier), Queensland yellowtail angelfish, Chaetodontoplus meredithi Kuiter, Pomacanthus imperator, P. semicirculatus, and P. sexstriatus. For the Ostraciidae, specimens were found in the yellow boxfish, Ostracion cubicum Linnaeus, and the spotted boxfish, Ostracion meleagris Shaw. cox1 sequences were generated for all host-parasite-locality combinations of the new samples from pomacanthids and ostraciids, and for a further eight known species of Paradiscogaster from other fishes (Table 2). Phylogenetic analysis of these data (Figure 2) resolved the species from pomacanthids and ostraciids into four principal clades. These four clades differed at 59–105 bp (p-distance = 12.45–22.15%). The first principal clade comprises specimens from Ostracion cubicum and O. meleagris from the Great Barrier Reef and Ningaloo Reef and forms three secondary clades (Lizard Island + Ningaloo Reef, Ningaloo Reef only and Heron Island) which differ at 8–15 bp (p-distance = 1.69–3.17%). Based on morphology, we interpret these clades as lineages of a single species, which is morphologically identified as Paradiscogaster ostracii (Yamaguti, 1934) Cribb, Anderson & Bray, 1999. The second principal clade resolved as sister to the third principal clade and likewise comprised specimens from Ostracion cubicum and O. meleagris, from Ningaloo Reef and Japan. Specimens of this principal clade differ at 1–23 bp (p-distance = 0.21–4.85%) but have no significant structuring in the cox1 analysis; these samples are morphologically identified as Paradiscogaster macrostomus (Shimazu & Kamegai, 1990) Cribb, Anderson & Bray, 1999. The third principal clade comprises specimens from Centropyge and Pomacanthus from the Great Barrier Reef and Ningaloo Reef; this clade resolved as the poorly supported sister to Paradiscogaster chaetodontis Yamaguti, 1938 within a larger clade of species mostly parasitic in chaetodontids. This principal clade forms three secondary clades (Heron Island + Lizard Island + Ningaloo Reef, Ningaloo Reef only, and Yonge Reef) which differ at 21–28 bp (p-distance = 4.43–5.91%) and are interpreted as representing three distinct species, Paradiscogaster machidai Cribb, Anderson & Bray, 1999 (Heron Island + Lizard Island + Ningaloo Reef) and two new species (Ningaloo Reef and Yonge Reef). These three species are almost morphologically cryptic, and delineation of them requires careful consideration of their morphology, molecular distinction, host range, and geographical distribution. The fourth principal clade comprises samples from C. meredithi and, based on morphology, is identified as Paradiscogaster leichhardti Diaz & Cribb, 2013. This species forms a well-supported clade with P. quasimodo Bray, Cribb & Barker, 1994, a species parasitic in chaetodontids.

Figure 2.

Phylogram from the unrooted neighbour-joining analysis of the cox1 mtDNA dataset for species that agree with the current generic diagnosis of Paradiscogaster Yamaguti, 1934. Bootstrap support values are shown at the nodes, with values of <85 not shown. The scale bar indicates the number of base differences.

Figure 2. Long description

A neighbor-joining phylogram with bootstrap values at nodes. The tree branches from a root on the left into several major clades.

* The top clade consists of Discogastroides ostracii specimens from various locations including Lizard Island, Ningaloo Reef, and Heron Island. Bootstrap support at the base of this clade is 95.

* Below this is a clade for Discogastroides macrostomus with specimens from Okinawa and Ningaloo Reef, supported by a bootstrap value of 99.

* The middle section contains several Paradiscogaster species including P. melendezi, P. munozae, P. glebulae, P. eniwetokensis, and P. oxleyi.

* A distinct clade follows for Paradiscogaster habilis with a bootstrap value of 99.

* The next major group includes Paradiscogaster chaetodontis followed by a large clade containing P. machidai, P. hoggettae, and P. ningalooensis, all supported by a bootstrap value of 99.

* The bottom of the tree features Crassitrema quasimodo and a final clade for Findovitellus leichhardti from Heron Island.

A scale bar at the bottom left indicates 10 base differences.

Specimens consistent with the current generic diagnosis of Parayamagutia were recovered from Ostracion cubicum and O. meleagris from off Lizard Island and Heron Island. cox1 data were generated for just two specimens, one from each location, which differ at 10 bp (p-distance = 2.11%). These specimens are morphologically identified as Parayamagutia ostracionis Machida, 1971, the sole representative of the genus.

Phylogenetic analyses of a comprehensive 28S dataset (not shown here), incorporating both the Gymnophalloidea and Microphalloidea, demonstrate the same dramatic paraphyly of the Faustulidae reported by numerous authors previously (e.g., Cutmore et al Reference Cutmore, Bray and Cribb2018; Sokolov et al Reference Sokolov, Ali, Adday, Othman and Vlasenkov2025; Sun et al Reference Sun, Bray, Yong, Cutmore and Cribb2014), with the new sequence data forming a strongly supported clade with species of the Zoogonidae, distant from the Faustulidae sensu stricto. Analysis of a dataset restricted to the Zoogonidae (Figure 3) demonstrates that the new samples from pomacanthids and ostraciids form a single, well-supported clade (together with some species from other fishes) within the Zoogonidae.

Figure 3.

Relationships between species of the Zoogonidae based on maximum likelihood analysis of the 28S dataset. Nodal support is indicated by a circle; relationships with strong support in both maximum likelihood and Bayesian inference analyses (>0.85 and >85, respectively) are indicated by filled circles, those with strong support in only the Bayesian inference analysis (>0.85) are indicated by unfilled circles. The scale-bar indicates expected number of substitutions per site.

Figure 3. Long description

A horizontal phylogenetic tree with a scale bar of 0.04 expected substitutions per site at the bottom left. The tree is rooted by an outgroup at the top left containing Renicola parvicaudatus, Nephromonorcha varitestis, and Renicola thapari.

Moving down the right side, the tree is organized into labeled clades:

* Cephaloporinae: Containing Plectognathotrema kamegaii.

* Lepidophyllinae: Containing Lepidophyllum steenstrupi and Lepidophyllum cameroni.

* Lecithostaphylinae: A large clade including Steganoderma cf. eamiqtrema, Proctophantastes gillissi, and two clusters of Deretrema, Innuptacola, and Lecithostaphylus species.

* Zoogoninae: Containing Diphterostomum sp., Pseudozoogonoides subaequiporus, Pseudozoogonoides ugui, and Zoogonoides viviparus.

* Antorchiinae: The largest clade at the bottom, containing numerous genera including Trigonocryptus, Findovitellus, Crassitrema, Neoparantorchis, Parantorchis, Scaticola, Parayamagutia, Discogastroides, and Paradiscogaster.

Nodes are marked with circles to indicate statistical support. Filled circles represent strong support in both maximum likelihood and Bayesian inference analyses. Unfilled circles represent strong support only in the Bayesian inference analysis. Branch lengths vary, reflecting the genetic distance between taxa.

For the specimens consistent with Antorchis, two of the three principal clades uncovered in the cox1 analysis are represented in the 28S analyses and resolved as sister (no 28S data are available for A. nasonis). Examination of morphological vouchers, informed by the phylogenetic topologies, revealed reliable morphological distinctions between the three principal clades of Antorchis. A re-examination of the reports of the eight known species of Antorchis [A. chaetodonis (Yamaguti, 1934) Machida, 1975, A. intermedius (Madhavi, 1975) Cribb, Anderson & Bray, 1999, A. lintoni Travassos, Artigas & Pereira, 1928, A. nasonis, A. parabotia Feng & Wang, 1997, A. pomacanthi, A. tsushimaensis (Machida, 1971) Machida, 1975, and A. urna (Linton, 1910) Linton, 1911] in light of the new phylogenetic analyses suggests that four genera are needed to accommodate species with this ‘Antorchis-like morphology’, for which we recognise Antorchis, Neoparantorchis Hafeezullah & Siddiqi, 1971, Parantorchis Yamaguti, 1934, and one new genus. A key for the delineation of the four genera is provided in the Discussion.

For the specimens consistent with Paradiscogaster, all four of the principal clades uncovered in the cox1 analysis are represented in the 28S analyses. The first and second principal clades in the cox1 analysis, identified as P. ostracii and P. macrostomus, respectively, form strongly supported sister clades in the 28S analyses. These two species then resolve as sister to Parayamagutia ostracionis. Together, these three species form a distinct and strongly supported clade of species parasitic in ostraciids which resolves as sister to the large clade of species of Paradiscogaster from chaetodontids, pomacanthids, and terapontids. Both P. ostracii and P. macrostomus differ from the type species of Paradiscogaster, and all other species of Paradiscogaster included in the phylogenetic analyses, in the position of the vitellarium (marginal at the level of the ventral sucker vs submarginal at the level of the caeca). On the basis of this topological and morphological distinction, we recognise Discogastroides Strand, 1934 for these two species. The third principal clade in the cox1 analysis, identified as Paradiscogaster machidai and two new species, resolves in the 28S analyses within a clade including seven other species of Paradiscogaster from chaetodontids and terapontids. These species are considered consistent with Paradiscogaster sensu stricto and are retained in that genus. The fourth principal clade in cox1, identified as P. leichhardti, forms a strongly supported clade with Paradiscogaster quasimodo in 28S analyses, distant from Paradiscogaster sensu stricto and sister to Trigonocryptus Martin, 1958; two new genera are proposed for these two species.

New sequence data for two species currently assigned to Bacciger, B. lesteri Bray, 1982 and B. sprenti Bray, 1982, resolve as the sister clade to the (Discogastroides + Parayamagutia) + Paradiscogaster clade. Both species were already considered incertae sedis in Bacciger by Curran et al (Reference Curran, Warren and Bullard2022), which is considered to belong to the Microphalloidea; a new genus is proposed for these species.

Based on the overall phylogenetic topology, we recognise the following genera as members of the Zoogonidae: Crassitrema n. g., Discogastroides, Findovitellus n. g., Neoparantorchis, Paradiscogaster, Parantorchis, Parayamagutia, Trigonocryptus, Nasoantorchis n. g., and Scaticola n. g. Based on morphology and host inferences enabled by the new phylogeny, we also transfer Antorchis sensu stricto to the Zoogonidae. A previously synonymised subfamily is available for this clade of zoogonids, the Antorchiinae Skrjabin & Koval, 1957.

Taxonomy

Family Zoogonidae Odhner, 1902

Subfamily Antorchiinae Skrjabin & Koval, 1957

Synonyms: Discogasterinae Yamaguti, 1934; Discogasteroidinae Srivastava, 1939; Antorchiinae Yamaguti, 1958.

Amended diagnosis

Body elongate or short, fusiform, pyriform, oval or conical with lateral folds. Tegument spinose. Oral sucker subglobular, infundibuliform or oval-pyriform, subterminal or terminal. Ventral sucker usually not strongly muscular, small to large, sometimes a large, muscular pad, in posterior two-thirds of body. Prepharynx short. Pharynx oval or subglobular, small. Oesophagus of variable length, slender. Intestinal bifurcation in forebody or dorsal to ventral sucker. Caeca typically short, rarely long, saccular, terminate in forebody or dorsal to ventral sucker. Testes subglobular, opposite, in forebody, hindbody or at level of ventral sucker. Cirrus-sac entirely in forebody or extends posteriorly to level of ventral sucker. Internal seminal vesicle bipartite. Pars prostatica long, narrow. Ejaculatory duct short. Genital pore median, rarely slightly sinistrally or dextrally submedial, in forebody. Ovary subglobular, submedial, at level of testes or dorsal to ventral sucker. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally. Uterine coils extensive in hindbody, sometimes extend into forebody, rarely almost entirely in forebody. Eggs tanned, operculate. Vitelline follicles in lateral clusters in forebody or at level of ventral sucker, comprising small to large number of small to large follicles. Excretory vesicle Y-shaped, occasionally V-shaped or subglobular. Excretory pore terminal. In intestine of marine fishes.

Type-genus: Antorchis Linton, 1911.

Other genera: Crassitrema n. g.; Discogastroides Strand, 1934; Findovitellus n. g.; Nasoantorchis n. g.; Neoparantorchis Hafeezullah & Siddiqi, 1971; Paradiscogaster Yamaguti, 1934; Parantorchis Yamaguti, 1934; Parayamagutia Machida, 1971; Scaticola n. g.; Trigonocryptus Martin, 1958.

Remarks

There are four previously synonymised subfamilies relating to genera considered here. Yamaguti (Reference Yamaguti1934) proposed the Discogasterinae Yamaguti, 1934 for Discogaster ostracii Yamaguti, 1934 and Discogaster minor Yamaguti, 1934. However, Discogaster Yamaguti, 1934 nec Burmeister, 1835 was a homonym, for which Discogastroides Strand, 1934 was proposed as a replacement. To accommodate the genus proposed by Strand (Reference Strand1934), Srivastava (Reference Srivastava1939) proposed the Discogasteroidinae Srivastava, 1939; however, this name was based on the misspelt Discogasteroides (instead of Discogastroides), and thus the subfamily name was also misspelt. The oldest available valid subfamily name is the Antorchiinae. The Antorchiinae was proposed twice in successive years, first by Skrjabin and Koval (Reference Skrjabin and Koval1957) and second by Yamaguti (Reference Yamaguti1958). Following principle of priority, we recognise Antorchiinae Skrjabin & Koval, 1957. Although we do not have sequence data for any species of Antorchis sensu stricto, the morphology and host associations of those species render the recognition of the Antorchiinae for all the presently considered taxa uncontroversial.

Antorchis Linton, 1911

Amended diagnosis

Body elongate, fusiform. Tegument spinose. Oral sucker infundibuliform or oval-pyriform, terminal, narrowing posteriorly. Ventral sucker not strongly muscular, in middle third of body. Prepharynx short. Pharynx oval, small. Oesophagus long, slender. Intestinal bifurcation in mid-forebody. Caeca short, saccular, thick-walled, terminate in forebody. Testes subglobular, opposite, in forebody or at level of anterior margin of ventral sucker. Cirrus-sac entirely in forebody or extends to just posterior to anterior margin of ventral sucker, distinctly recurved to form inverted U-shape. Internal seminal vesicle bipartite. Pars prostatica narrow, ensheathed by conspicuous prostatic cells. Genital pore median, immediately anterior to ventral sucker. Ovary subglobular, post-testicular, submedial, at level of ventral sucker. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally at conspicuous opening. Uterine coils extensive in hindbody, rarely extend into forebody past level of testes. Eggs tanned, operculate. Vitelline follicles in two lateral clusters in forebody, comprising moderate to large number of small to large follicles. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of Atlantic pomacanthid fishes.

Type species: Antorchis urna (Linton, 1910) Linton, 1911.

Other species: Antorchis holacanthi Siddiqi & Cable, 1960.

Species inquirenda: Antorchis lintoni Travassos, Artigas & Pereira, 1928; Antorchis parabotia Feng & Wang, 1997.

Remarks

Antorchis was proposed by Linton (Reference Linton1911) as a replacement name for the homonym Mesorchis Linton, 1910 nec Dietz, 1909, proposed just a year prior for Mesorchis urna Linton, 1910. Prior to our analysis, eight species were recognised in the genus: six species from marine fishes of the tropical Indo-West Pacific, Caribbean and Gulf of Mexico (A. chaetodonis, A. intermedius, A. nasonis, A. pomacanthi, A. tsushimaensis, and A. urna), and two from freshwater fishes (A. lintoni from South America and A. parabotia from Asia). We have assigned the five species from the Indo-West Pacific (A. chaetodonis, A. intermedius, A. nasonis, A. pomacanthi, and A. tsushimaensis) to new genera, leaving three in Antorchis: A. urna, A. lintoni, and A. parabotia. In addition to these three species, we here consider the previously synonymised species Antorchis holacanthi Siddiqi & Cable, 1960 as valid. Antorchis urna and A. holacanthi both possess a distinctly recurved cirrus-sac, a genital pore at the anterior margin of the ventral sucker, and infect pomacanthid fishes; thus, we consider both valid members of the genus. Based on differences in morphology, host identity, and ecology, we think it unlikely that A. lintoni and A. parabotia are true members of Antorchis, and the two species are here considered species inquirendae.

Antorchis urna (Linton, 1910) Linton, 1911

Synonyms: Mesorchis urna Linton, 1910.

Type-host: Pomacanthus arcuatus (Linnaeus), Gray angelfish (Acanthuriformes: Pomacanthidae).

Other hosts: Holacanthus bermudensis Goode, Bermuda blue angelfish; Holacanthus ciliaris (Linnaeus), Queen angelfish; Pomacanthus paru (Bloch), French angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Dry Tortugas, Florida.

Other localities: Bahamas; Drowned Cays, Belize; Cuba; Curacao; Biscayne Bay, Florida; Jamaica.

Remarks

Linton (Reference Linton1910) described Antorchis urna (as Mesorchis urna Linton, 1910) from two pomacanthid species, the Bermuda blue angelfish, Holacanthus bermudensis Goode (as Angelichthys isabelita Jordan & Rutter), and the gray angelfish, Pomacanthus arcuatus (Linnaeus) from off Florida; in his description, he did not nominate either species as the type-host. Since its description, this species has been reported from a range of sites in the Caribbean: from P. arcuatus from off Belize (Fischthal Reference Fischthal1977), Cuba (Vigueras Reference Vigueras1955), Florida (Overstreet Reference Overstreet1969), Jamaica and Curacao (Nahhas and Cable Reference Nahhas and Cable1964), Mexico (Caballero-Rodríguez et al Reference Caballero-Rodríguez, Del Río, Rodríguez and Tello1992) and Puerto Rico (Dyer et al Reference Dyer, Williams and Bunkley-Williams1985, Reference Dyer, Williams and Bunkley-Williams1992, Reference Dyer, Williams and Bunkley-Williams1998; Siddiqi and Cable Reference Siddiqi and Cable1960), from the French angelfish, Pomacanthus paru (Bloch), from the Bahamas (Sogandares-Bernal Reference Sogandares-Bernal1959), and Florida (Overstreet, Reference Overstreet1969), from H. bermudensis from Florida (Overstreet Reference Overstreet1969) and from the queen angelfish, H. ciliaris (Linnaeus), off Belize (Fischthal Reference Fischthal1977). Because Linton did not nominate either pomacanthid species as the type-host, the two specimens he lodged (USNM 1321227 and 1321228) are considered syntypes. Linton’s illustration of this species was of a sample from Pomacanthus arcuatus, and we thus designate that specimen (USNM 1321228) as the lectotype.

Antorchis holacanthi Siddiqi & Cable, 1960

Type-host: Holacanthus tricolor (Bloch), Rock beauty (Acanthuriformes: Pomacanthidae).

Other host: Holacanthus ciliaris (Linnaeus), Queen angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: La Parguera, Puerto Rico.

Other locality: Jamaica.

Remarks

Siddiqi and Cable (Reference Siddiqi and Cable1960) described Antorchis holacanthi from two species of pomacanthid, the rock beauty, Holacanthus tricolor (Bloch), and Holacanthus ciliaris, from off Puerto Rico; this species has since been reported from H. tricolor from Jamaica by Nahhas and Cable (Reference Nahhas and Cable1964) and Nahhas and Carlson (Reference Nahhas and Carlson1994). Antorchis holacanthi was differentiated from A. urna by body shape and size, oral sucker shape, pharynx size, position of the testes, extent of the uterus, and, most importantly, the form of the vitellaria (few large follicles for Antorchis holacanthi vs many small follicles for A. urna). Overstreet (Reference Overstreet1969) considered the difference between A. holacanthi and A. urna within the range of intraspecific variation and, given the related hosts and location, synonymised the two species, though nonetheless commented on the difference in the vitelline follicles. Nahhas and Carlson (Reference Nahhas and Carlson1994) detected the same difference in vitelline follicles in samples from pomacanthids of Jamaica, and, disagreeing with the synonymy of Overstreet (Reference Overstreet1969), recognised both A. holacanthi and A. urna as valid. Based on integrated molecular and morphological analysis of our new samples from the tropical Indo-West Pacific, we conclude that differences in vitelline follicle form and number can indeed be a reliable morphological distinction and recognise A. holacanthi as a valid species.

Given that Antorchis holacanthi is known only from species of Holacanthus, and that all reports of Antorchis urna from species of Holacanthus were not illustrated, we examined the available lodged material identified as Antorchis urna from species of Holacanthus [USNM 1321227 of Linton (Reference Linton1910) and USNM 1369748 of Fischthal (Reference Fischthal1977)]. Despite the forebody of Linton’s sample being partially destroyed, we observed that both specimens had many small vitelline follicles rather than few large follicles, consistent with their identification as A. urna.

Antorchis lintoni Travassos, Artigas & Pereira, 1928 species inquirenda

Type-host: Psalidodon fasciatus (Cuvier), Banded astyanax (Characiformes: Characidae).

Type-locality: São Paulo, Brazil.

Remarks

Antorchis lintoni was described by Travassos et al (Reference Travassos, Artigas and Pereira1928) from a freshwater characid, the banded astyanax Psalidodon fasciatus (Cuvier), from Brazil. The description was based on a single specimen and lacked information critical for an accurate designation; Travassos et al (Reference Travassos, Artigas and Pereira1928) even noted in their description that the extensive uterus made examination of the single specimen difficult and that their identification was made with reservations. Based on the lack of detail in the description (especially regarding the cirrus-sac), the inconsistent host (freshwater as opposed to marine) and some observable morphological differences relative to all other Antorchis species (e.g., no oesophagus vs long oesophagus), we think A. lintoni requires transfer elsewhere; however, currently it is unclear to which genus, or even family, it should be moved.

Antorchis parabotia Feng & Wang, 1997 species inquirenda

Type-host: Parabotia banarescui (Nalbant), loach (Cypriniformes: Cobitidae).

Type-locality: Central China.

Remarks

Antorchis parabotia was described by Feng and Wang (Reference Feng, Wang and Song1997) from a freshwater cobitid, the loach, Parabotia banarescui (Nalbant), from central China. This species does resemble species of Antorchis in having short caeca that do not exceed the testes, a recurved cirrus-sac, and the genital pore close to the anterior margin of the ventral sucker. Despite these morphological similarities, given the distinction in host (Cypriniformes vs Acanthuriformes), and geographical location (freshwater Asia vs marine Caribbean), we think A. parabotia requires transfer to a more appropriate group; however, until molecular data are generated for this species, it is unclear to where it should be moved.

Parantorchis Yamaguti, 1934

Amended diagnosis

Body fusiform or oval. Tegument spinose. Oral sucker subglobular, subterminal. Ventral sucker not strongly muscular, in middle or posterior third of body. Prepharynx short. Pharynx oval, small. Oesophagus moderate to long, slender. Intestinal bifurcation in mid-forebody. Caeca medium length, saccular, thick-walled, terminate posterior to testes or at level of posterior margin of testes, always extend posteriorly beyond middle of testes. Testes subglobular, opposite, in forebody or at level of anterior margin of ventral sucker. Cirrus-sac entirely in forebody or extends to just posterior to anterior margin of ventral sucker, straight to slightly curved. Internal seminal vesicle bipartite. Pars prostatica narrow, ensheathed by conspicuous prostatic cells. Eversible ejaculatory duct short. Genital pore median, at level of intestinal bifurcation. Ovary subglobular, pre-testicular or intratesticular at level of anterior testicular margin, submedial, anterior to ventral sucker. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally. Uterine coils extensive posterior to testes. Eggs tanned, operculate. Vitelline follicles in two lateral clusters in forebody, each cluster comprising a small number (<15) of medium-sized follicles. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of chaetodontid and pomacanthid fishes.

Type-species: Parantorchis chaetodonis Yamaguti, 1934.

Other species: Parantorchis intermedius Madhavi, 1975; Parantorchis vaili n. sp.; Parantorchis parviovatus n. sp.

Remarks

Yamaguti (Reference Yamaguti1934) proposed Parantorchis for Parantorchis chaetodonis Yamaguti, 1934, collected from an unidentified Chaetodon species (Chaetodontidae) from southern Japan. Yamaguti (Reference Yamaguti1934) remarked that while it was similar to species of Antorchis, it differed in the position of the ovary (pre-testicular vs post-testicular). Two further species were erected in this genus, Parantorchis intermedius Madhavi, 1975 and Parantorchis tsushimaensis Machida, 1971, before it was synonymised with Antorchis by Machida (Reference Machida1975). Our phylogenetic analyses demonstrate that specimens with caeca that extend past the testes, have few vitelline follicles, and the ovary pre-testicular or intratesticular at the level of the anterior testicular margin are genetically distinct from those with caeca that do not extend past the testes, have many small vitelline follicles, and a more posterior ovary; we recognise Parantorchis for species with the former morphology and Neoparantorchis for species with the latter morphology (see below).

Parantorchis chaetodonis Yamaguti, 1934

Synonyms: Antorchis chaetodonis (Yamaguti, 1934) Machida, 1975; Antorchis chaetodontis (Yamaguti, 1934) Machida, 1975.

Type-host: Chaetodon sp., unknown butterflyfish (Acanthuriformes: Chaetodontidae).

Type-locality: Kyūshū, Japan.

Remarks

Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) detailed the issues surrounding this species, including questions of host identity (chaetodontid vs pomacanthid), morphological variation between the holotype and paratypes, and rarity (not reported since its original description). While this makes the species enigmatic and worthy of further study, in our opinion it does not detract from its validity. A second species of this genus, Parantorchis intermedius Madhavi, 1975, is known only from a chaetodontid. A great number, perhaps the majority, of marine trematodes in the tropical Indo-West Pacific have rarely been reported since their description (Cribb et al Reference Cribb, Bray, Diaz, Huston, Kudlai, Martin, RQ-Y and Cutmore2016; Poulin Reference Poulin2025), and questions of multiple species within the type series are certainly not exceptional, with the occurrence of morphologically similar congeners found in the same host individual fish becoming a seemingly regular occurrence (e.g., Bray et al. Reference Bray, Cutmore and Cribb2023; Cribb et al. Reference Cribb, Cutmore, Wee, Browne, Diaz and Pitt2024; Cutmore et al Reference Cutmore, Corner and Cribb2023; Wee et al Reference Wee, Cutmore and Cribb2019). All these problems would benefit from further study of chaetodontids and pomacanthids from the type location of this species. Abdel-Mawla and Yousef (Reference Abdel-Mawla and Yousef2019) recently reported this species from the Red Sea; however, based on the host, Scomberomorus commerson (Lacépède) and the photomicrograph provided in the report, it appears that the specimens in that report are more likely hemiurids.

Parantorchis intermedius Madhavi, 1975

Synonyms: Antorchis intermedius (Madhavi, 1975) Cribb, Anderson & Bray, 1999.

Type-host: Chaetodon pictus Forsskål, Horseshoe butterflyfish (Acanthuriformes: Chaetodontidae).

Type-locality: Bay of Bengal, India.

Remarks

This species was described on two specimens from a single host individual and has not been reported since. As stated above, the infection of a chaetodontid, rather than a pomacanthid, is unusual for this assemblage of genera (Antorchis, Parantorchis, and Neoparantorchis) but broadly plausible given that both families are heavily reef-associated and are in the Acanthuriformes. Additionally, given the distinctive appearance of Chaetodon pictus, it seems very unlikely that the host identification was wrong.

Parantorchis vaili n. sp. (Figure 4a)

Figure 4.

New species of Parantorchis Yamaguti, 1934 ex pomacanthids from the northern Great Barrier Reef, Australia. (a) Parantorchis vaili n. sp. ex Pomacanthus sexstriatus (Cuvier) from off Lizard Island, ventral view; (b) Parantorchis parviovatus n. sp. ex Pomacanthus imperator (Bloch) from Yonge Reef, ventral view. Scale bars = 400 μm.

Figure 4. Long description

Two anatomical line drawings of flatworms.

Panel a: Parantorchis vaili. At the anterior end is a large circular oral sucker. Below this, a narrow esophagus leads to a central region containing a cirrus sac and a large, circular ovary. Flanking the mid-body are clusters of dark, rounded vitelline follicles. The posterior half is densely packed with numerous small, oval eggs. A vertical scale bar to the right represents 400 micrometers.

Panel b: Parantorchis parviovatus. This specimen is broader and more heart-shaped than the first. It features a similar anterior oral sucker and central reproductive organs, including a prominent cirrus sac and ovary. The vitelline follicles are clustered in the upper lateral regions. The uterus is highly coiled and filled with eggs, extending throughout the lower two-thirds of the body. A vertical scale bar to the right represents 400 micrometers.

Type-host: Pomacanthus sexstriatus (Cuvier), Sixbar angelfish (Acanthuriformes: Pomacanthidae).

Other host: Pomacanthus imperator (Bloch), Emperor angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Off Lizard Island (14°40′S, 145°27′E), northern Great Barrier Reef, Queensland, Australia.

Other locality: Yonge Reef (14°35′S, 145°37′E), northern Great Barrier Reef, Queensland, Australia.

Site in host: Intestine.

Prevalence: Lizard Island: P. sexstriatus, 1/3 (33%). Yonge Reef, P. imperator: 1/1 (100%).

Deposition of specimens: Holotype (G242325), 25 paratype specimens (G242326–50) and 29 voucher specimens (G242351–79) lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518168–70; ITS2 rDNA, PZ510920–21; 28S rDNA, PZ513251.

ZooBank LSID: urn:lsid:zoobank.org:act:24380B55-3682-4992-B8A3-EC51961BEAEF.

Etymology: This species is named for Dr Lyle Vail, who, with his wife, Dr Anne Hoggett, is the recently retired joint Director of the Lizard Island Research Station. We name this species for Lyle in appreciation of a lifetime’s contribution to fostering, facilitating and promoting Australian marine science.

Description

[Measurements based on 20 wholemounts and 6 hologenophore partial mounts.] Body fusiform, distinctly pointed posteriorly, widest at level of ventral sucker, 930–1,105 × 302–405 (1,023 × 358); body length to width ratio 2.50–3.45 (2.86):1. Tegument spinose; spines smaller in hindbody. Forebody 492–589 (545) long, occupying 51–57 (53)% of body length. Oral sucker subspherical, 106–130 × 133–165 (118 × 144). Ventral sucker roughly round, sometimes with distinctly straight anterior margin, not strongly muscular, without evident cavity, in posterior half of body, 131–158 × 122–165 (143 × 139); oral sucker length to ventral sucker length ratio 0.77–0.90 (0.83):1; oral sucker width to ventral sucker width ratio 0.93–1.14 (1.05):1. Prepharynx short, always within posterior cavity of oral sucker. Pharynx small, oval, 41–55 × 51–70 (47 × 59); pharynx length to oral sucker length ratio 0.36–0.46 (0.40):1; pharynx width to oral sucker width ratio 0.35–0.44 (0.41):1. Oesophagus long, straight or slightly sinuous, 127–230 (179) long. Intestinal bifurcation in posterior half of forebody. Caeca medium length, saccular, reach posteriorly to at least middle of testes, often extending past testes, terminating at level of ventral sucker; left caecum 241–368 (298) long; right caecum 248–368 (298) long; caecal field 217–333 (265) long, occupying 21–35 (26)% of body length; caecal length to oesophagus length 1.17–2.79 (1.69):1.

Testes two, ellipsoidal to round, entire, opposite, in forebody or at level of ventral sucker; left testis 80–114 × 64–92 (98 × 83); right testis 86–128 × 59–99 (100 × 83); post-testicular space 414–503 (450) long, occupying 41–51 (44)% of body length. Cirrus-sac extends anteriorly from level of anterior margin of ventral sucker to common genital pore, with posterior end variably dextral or sinistral to midline, 141–269 × 85–216 (193 × 113). Internal seminal vesicle bipartite; anterior portion saccular, 29–94 × 28–94 (46 × 40); posterior portion tubular. Pars prostatica straight, prominent. Eversible ejaculatory duct short. Genital atrium distinct. Genital pore median, at level of intestinal bifurcation, 320–406 (351) from anterior margin, 602–770 (677) from posterior margin.

Ovary in forebody, entire, smaller than, and between, testes, submedial, opposite cirrus-sac, 69–104 × 44–89 (78 × 62); post-ovarian space 433–542 (479) long. Canalicular seminal receptacle present, medial, posterior to ovary, dorsal to ventral sucker. Laurer’s canal opens dorsally, in the middle third of the hindbody. Dorsal invagination prominent, distinctly anterior to opening of Laurer’s canal; associated glandular cells present. Vitelline follicles in two separate lateral fields, globular, 28–49 (36) in diameter; 17–22 (20) follicles in total; 7–12 (10) follicles in left field; 7–13 (10) follicles in right field; vitelline field extends to near level of anterior margin of testes to just anterior to genital pore, 132–202 (162) long, occupying 13–22 (16)% of body length; 216–333 (292) from anterior extremity; 516–652 (577) from posterior extremity. Uterus extensive posterior to testes, forms distinct V-shape, with portion leading to genital atrium ascending from centre of V. Eggs tanned, operculate, 33–45 × 18–24 (38 × 20).

Excretory pore terminal. Excretory vesicle Y-shaped; stem short; arms terminate at level of ventral sucker.

Remarks

Specimens of this new species were found in just one of five Pomacanthus sexstriatus and one of four Pomacanthus imperator examined during this study. We re-examined the voucher specimens of Neoparantorchis pomacanthi lodged in the Queensland Museum by Cribb et al (Reference Cribb, Anderson and Bray1999) and Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) and examined all their unpublished material of this species (currently held in the Queensland Museum), and found no evidence of this species in a further 10 Pomacanthus sexstriatus examined as part of those studies. Notably, no specimens of Neoparantorchis pomacanthi were found in the hosts that were infected with Parantorchis vaili; the reason behind the exclusivity of infections by these two species is unknown.

Parantorchis vaili is readily distinguished from Parantorchis chaetodonis in having caeca that reach only to the posterior margin of the testes (vs well past the testes), larger oral sucker to ventral sucker length and width ratios (length ratio 0.77–0.90:1 vs 0.58:1; width ratio 0.93–1.14:1 vs 0.53:1), and a smaller body (930–1,105 × 302–405 vs 1,580 × 1,000), smaller oral sucker (106–130 × 133–165 vs 190 × 210), smaller ventral sucker (131–158 × 122–165 vs 330 × 400), smaller pharynx (41–55 × 51–70 vs 85 × 95), smaller testes (left testis 80–114 × 64–92 vs 190 × 180; right testis 86–128 × 59–99 vs 210 × 180), smaller cirrus-sac (141–269 × 85–216 vs 320 × 190) and smaller ovary (69–104 × 44–89 vs 150 × 140). These distinctions are made in regard to the holotype data reported by Yamaguti (Reference Yamaguti1934) rather than the expanded data reported by Machida (Reference Machida1975), given the possibility of the type-series representing multiple species, as noted by Machida (Reference Machida1975) and Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015).

Parantorchis vaili is morphologically highly similar to P. intermedius; the two species differ only slightly in the size of the cirrus-sac (141–269 vs 312–390 long), the overall body shape (the greatest width is at the level of the ventral sucker for P. vaili and more anteriorly at the level of the testes for P. intermedius), and the pharynx to oral sucker ratio (with P. vaili having a comparatively larger pharynx). However, the most important difference between these species is arguably in the host identity (pomacanthids vs chaetodontids). Parantorchis vaili has been found only in Pomacanthus species at a location where hundreds of individuals of 26 species of chaetodontids have been examined, whereas P. intermedius is known only from a chaetodontid. To date, we know of no individual trematode species that has been definitively shown to infect fishes of both families, despite their common co-occurrence. There are two species that have been reported from fishes of both families, but both are problematic. Paragyliauchen chaetodontis Yamaguti, 1934 (Gyliauchenidae) was described from an unidentified species of Chaetodon by Yamaguti (Reference Yamaguti1934) but later reported exclusively from pomacanthids: from the bluestriped angelfish, Chaetodontoplus septentrionalis (Temminck & Schlegel), by Ozaki (Reference Ozaki1937), Machida et al (Reference Machida, Ichihara and Kamegai1970) and Machida and Uchida (Reference Machida and Uchida2001), and from the blackstriped angelfish, Genicanthus lamarck (Lacepède), and the spotbreast angelfish, Genicanthus melanospilos (Bleeker), by Machida and Uchida (Reference Machida and Uchida2001). We consider the identification of Chaetodon sp. suspect, and this pattern of host-specificity requires confirmation. Thulinia microrchis (Yamaguti, 1934) Bray, Cribb & Barker, 1993 (Bunocotylidae) was reported from the citron butterflyfish, Chaetodon citrinellus Cuvier, Klein’s butterflyfish, Chaetodon kleinii Bloch, and Chaetodontoplus meredithi by Bray et al (Reference Bray, Cribb and Barker1993). This host range is not supported by molecular data; however, given the notoriously low host-specificity of hemiuroids, even demonstrated with molecular data (e.g., Bouguerche et al Reference Bouguerche, Huston, Karlsbakk, Ahmed and Holovachov2024; Krupenko et al Reference Krupenko, Gonchar, Krapivin, Kremnev and Skobkina2025), the report of T. microrchis from both fish families is reasonable, but like other hemiuroids, the specificity should be treated as an exception rather than the rule. On balance, we consider the apparent host difference between P. intermedius and P. vaili to be a decisive distinction.

Parantorchis parviovatus n. sp. (Figure 4b)

Type-host: Pomacanthus imperator (Bloch), Emperor angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Yonge Reef (14°35′S, 145°37′E), northern Great Barrier Reef, Queensland, Australia.

Site in host: Intestine.

Prevalence: 1/1 (100%).

Deposition of specimens: Holotype (G242380) lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518167; 28S rDNA, PZ513250.

ZooBank LSID: urn:lsid:zoobank.org:act:D3A4725E-0AFB-44B2-A74B-A87E23A1F3FC.

Etymology: This species is named for size of the eggs, from the Latin parvus, meaning little, and ovum, meaning egg.

Description

[Based on a single hologenophore partial mount.] Body fusiform, distinctly pointed posteriorly, widest at level of anterior margin of ventral sucker, body width 569. Tegument spinose; spines prominent in forebody, sparser and smaller in hindbody. Ventral sucker roughly round, not strongly muscular, without evident cavity, 187 × 206, in posterior half of body, 344 from posterior margin. Oesophagus long. Intestinal bifurcation in posterior half of forebody. Caeca medium length, saccular, reach posteriorly to posterior margin of testes; left caecum 316 long; right caecum 292 long; caecal field 241 long.

Testes two, ellipsoidal, entire, opposite, in forebody, immediately anterior to ventral sucker; left testis 85 × 69; right testis 76 × 59; post-testicular space 532 long. Cirrus-sac extends anteriorly from level of testes to common genital pore, enters common genital atrium dorsally, 248 × 123. Internal seminal vesicle bipartite; anterior portion saccular, 104 × 98; posterior portion tubular. Pars prostatica straight, prominent. Eversible ejaculatory duct short. Genital atrium distinct, 19 in diameter. Genital pore median, at level of intestinal bifurcation, 761 from posterior margin of body.

Ovary in forebody, dextrally submedial, entire, larger than testes, 97 × 88; post-ovarian space 572 long. Canalicular seminal receptacle present, medial, dorsal to medial margins of ovary and cirrus-sac. Laurer’s canal opens dorsally in middle third of hindbody. Dorsal invagination and associated glandular cells not observed. Vitelline follicles in two separate lateral fields, globular, 37–39 (38) in diameter; 12 follicles in left field; 9 follicles in right field. Vitelline field extends near level of anterior margin of testes to just anterior to genital pore, 184 long, 572 from posterior extremity. Uterus extensive posterior to testes, forms V-shape, with portion leading to genital atrium ascending from centre of V. Eggs tanned, operculate, 18–22 × 10–12 (20 × 11).

Excretory pore terminal. Excretory vesicle Y-shaped; stem short, thin; arms broad, terminate at level of testes.

Remarks

Our decision to formally describe this species based on a single hologenophore sample was carefully considered and is not something we normally advocate; we certainly prefer to base descriptions on multiple specimens (at least 10–15 worms), preferably collected from multiple host individuals. We did, however, choose to formally describe this single specimen as it is unequivocally distinct, both morphologically and phylogenetically, and the chances of obtaining more specimens appear to be low.

The new species is clearly morphologically distinct in the size of the eggs. The eggs of P. parviovatus measure just 18–22 μm long. These are substantially smaller than those of P. vaili (33–45 μm), P. chaetodonis (29–36 μm), and P. intermedius (35–39 μm), and are smaller than those of almost all described antorchiine zoogonids. Of these zoogonids, only two species are reported to have similar sized or smaller eggs; Paradiscogaster hainanensis (Shen, 1990) Cribb, Anderson & Bray, Reference Cribb, Anderson and Bray1999 (from a monacanthid) has similar sized eggs (18–21 μm) and Paradiscogaster niger Bilqees, Reference Bilqees1976 (from a carangid) is reported as having minute eggs (4–10 μm). The minute eggs reported by Bilqees (Reference Bilqees1976) are so small relative to the rest of the group that the report seems implausible and requires further investigation. Parantorchis parviovatus further differs from P. chaetodonis in being generally smaller (having a smaller body width, ventral sucker, testes, cirrus-sac and ovary) and having caeca that reach to the posterior margin of the testes (vs well past the testes), and from P. intermedius in having a larger ventral sucker, shorter caeca, and smaller testes and cirrus-sac. Parantorchis parviovatus is most similar to P. vaili, but further differs in having a greater body width (569 vs 302–405) and a larger ventral sucker (187 × 206 vs 131–158 × 122–165).

The new specimen is clearly genetically distinct from other sequenced taxa, forming a clade with Parantorchis vaili in all analyses, but differing from it at 35–36 bp (p-distance = 7.38–7.60%) in the cox1 dataset and 16 bp (p-distance = 1.21%) in the 28S dataset; ITS2 data could not be generated for this species. Such differences, particularly those for the 28S region, are consistent with species level distinctions in other zoogonid genera as well as many other trematode genera found in the region.

We suspect that it will take significant effort to obtain more samples of this species. Although we have rarely actively targeted pomacanthids, we have surveyed trematodes of the tropical Indo-West Pacific broadly, and sampled pomacanthids whenever they have been encountered. We have sampled over 20,000 fishes from the region over the past 35 years, and in that time have only managed to examine four Pomacanthus imperator, one from French Polynesia, two from Ningaloo Reef, and one from the Great Barrier Reef. Of these four hosts, just one was infected by this species, and from that one fish just a single specimen was collected. In the context of a dwindling taxonomic capacity globally, and an imperative to describe species before they are lost along with their hosts, we think it best to describe distinctive species if the distinction is unambiguous.

Neoparantorchis Hafeezullah & Siddiqi, 1971

Amended diagnosis

Body elongate, fusiform or short, pyriform. Tegument spinose. Oral sucker subglobular, subterminal. Ventral sucker not strongly muscular, in posterior third of body. Prepharynx short. Pharynx oval, small. Oesophagus moderate to long, slender. Intestinal bifurcation in mid-forebody. Caeca short, saccular, thick-walled, terminate anterior to testes or at level of anterior margin of testes, never extend posteriorly beyond middle of testes. Testes subglobular, opposite, in forebody or at level of anterior margin of ventral sucker. Cirrus-sac entirely in forebody or extends to just posterior to anterior margin of ventral sucker, straight to slightly curved. Internal seminal vesicle bipartite. Pars prostatica narrow, ensheathed by conspicuous prostatic cells. Eversible ejaculatory duct short. Genital pore median, at level of intestinal bifurcation. Ovary subglobular, post-testicular or intertesticular at level of posterior testicular margin, submedial, anterior to ventral sucker. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally at conspicuous opening. Uterine coils extensive posterior to testes. Eggs tanned, operculate. Vitelline follicles in two lateral clusters in forebody, each comprising moderate to large number (>15) of small follicles. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of Indo-West Pacific pomacanthid fishes.

Type-species: Neoparantorchis pomacanthi (Hafeezullah & Siddiqi, 1970) Hafeezullah & Siddiqi, 1971.

Other species: Neoparantorchis tsushimaensis (Machida, 1971) n. comb.

Remarks

Neoparantorchis was proposed by Hafeezullah and Siddiqi (Reference Hafeezullah and Siddiqi1971) as a replacement name for the homonym Jonesiella Hafeezullah & Siddiqi, 1970 nec Brady, 1880, which was proposed a year earlier for Jonesiella pomacanthi Hafeezullah & Siddiqi, 1970. In the original description of Jonesiella, Hafeezullah and Siddiqi (Reference Hafeezullah and Siddiqi1970) noted that the genus was most similar to Parantorchis but differed in the possession of a post-testicular ovary (vs pre-testicular), a bipartite seminal vesicle (vs unipartite), a posteriorly positioned ventral sucker (vs medially positioned) and short caeca that do not extend posteriorly past the testes (vs extending beyond the testes posteriorly). Notably, they did not distinguish the new genus from Antorchis. In a review of the group, Machida (Reference Machida1975) noted that these morphological features agreed with the diagnosis of Antorchis and synonymised Neoparantorchis (and Parantorchis) with Antorchis. Based on integrated analyses of the new specimens, we agree with Hafeezullah and Siddiqi (Reference Hafeezullah and Siddiqi1970) that the differences outlined distinguish forms at the generic, rather than specific, level, and we consider Neoparantorchis as distinct from Antorchis. In addition to Neoparantorchis pomacanthi, we here suggest that one other known species belongs to the genus, and propose Neoparantorchis tsushimaensis (Machida, Reference Machida1971) n. comb. Both species have caeca that do not extend posteriorly past the testes, the genital pore at the level of the intestinal bifurcation, and the ovary post-testicular or intertesticular at the level of the posterior testicular margin.

Neoparantorchis pomacanthi (Hafeezullah & Siddiqi, 1970) Hafeezullah & Siddiqi, 1971

Synonyms: Jonesiella pomacanthi Hafeezullah & Siddiqi, 1970; Antorchis pomacanthi (Hafeezullah and Siddiqi, 1970) Machida, 1975; Parantorchis pomacanthi (Hafeezullah and Siddiqi, 1970) Madhavi, 1975.

Type-host: Pomacanthus annularis (Bloch), Bluering angelfish (Acanthuriformes: Pomacanthidae).

Other hosts: Pomacanthus semicirculatus (Cuvier), Blue angelfish; Pomacanthus sexstriatus (Cuvier), Sixbar angelfish; Pomacanthus xanthometopon (Bleeker), Yellowface angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Off Waltair, India.

Other localities: Off Lizard Island, Queensland, Australia; off Heron Island, Queensland, Australia; Ningaloo Reef, Western Australia; New Caledonia; Okinawa, Japan.

New material

New host: Pomacanthus imperator (Bloch), Emperor angelfish (Acanthuriformes: Pomacanthidae).

Known hosts: Pomacanthus semicirculatus; Pomacanthus sexstriatus; Pomacanthus xanthometopon.

Known localities: Off Lizard Island (14°40′S, 145°27′E), northern Great Barrier Reef, Queensland, Australia. Ningaloo Reef (23°12′S 113°45′E), Western Australia.

Site in host: Intestine.

Prevalence: Lizard Island: Pomacanthus semicirculatus, 1/1 (100%); Pomacanthus sexstriatus, 1/3 (33%); Pomacanthus xanthometopon, 1/1 (100%). Ningaloo Reef: Pomacanthus imperator, 1/2 (50%); Pomacanthus semicirculatus, 1/3 (33%); Pomacanthus sexstriatus, 2/2 (100%).

Deposition of specimens: 111 voucher specimens (G242381–491) lodged in the QM; 23 voucher specimens (V13655–77), including hologenophores, lodged in the WAM.

Molecular sequence data: cox1 mtDNA, PZ518149–65; ITS2 rDNA, PZ510916–19; 28S rDNA, PZ513247–49.

Remarks

Neoparantorchis pomacanthi was described by Hafeezullah and Siddiqi (Reference Hafeezullah and Siddiqi1970) from Pomacanthus annularis from India, and has since been reported several times, primarily from Australian waters. It has been reported from Pomacanthus semicirculatus from off Heron Island (Cribb et al Reference Cribb, Anderson and Bray1999; Reference Cribb, Bray, Hall and Cutmore2015), Ningaloo Reef (Cribb et al Reference Cribb, Bray, Hall and Cutmore2015) and Japan (Machida and Uchida Reference Machida and Uchida2001), from Pomacanthus sexstriatus from off Heron Island (Cribb et al Reference Cribb, Anderson and Bray1999, Reference Cribb, Bray, Littlewood, Pichelin, Herniou, Littlewood and Bray2001, Reference Cribb, Bray, Hall and Cutmore2015; Olson et al Reference Olson, Cribb, Tkach, Bray and Littlewood2003), Lizard Island and New Caledonia (Cribb et al Reference Cribb, Bray, Hall and Cutmore2015), and from Pomacanthus xanthometopon from off Lizard Island (Cribb et al Reference Cribb, Bray, Hall and Cutmore2015).

The cox1 data for the new specimens of N. pomacanthi formed four clades consistent with collection locality (one clade from each of the Great Barrier Reef and New Caledonia, and two clades at Ningaloo Reef) that differed at 25–33 bp (p-distance = 5.27–6.96%); minor intraclade variation was found at 0–5 bp (p-distance = 0–1.06%). Three of these clades are identical in the ITS2 dataset, and the fourth differs at just 1 bp (p-distance = 0.23%). Notably, however, these four cox1 clades differed at 2–9 bp (p-distance = 0.15–0.68%) in the 28S dataset. Two aspects of these data are significant. First, the distinction in the cox1 clades is substantial; 5.27–6.96% between clades is greater than the intraspecific variation recognised than for any other combination of species within this study and is close to that between some morphologically distinct species. However, these N. pomacanthi clades represent specimens separated by substantial geographic distances, although the presence of two clades at Ningaloo Reef is noteworthy. Second, variation of up to 9 bp in 28S is unusually high to be interpreted as intraspecific for fish trematodes in the tropical Indo-West Pacific, including cases considered over geographic range, with species delineated despite lower levels of variation for other families (e.g., Bray et al Reference Bray, Cutmore and Cribb2018; Cribb et al Reference Cribb, Bray, Justine, Reimer, Sasal, Shirakashi and Cutmore2022; Huston et al Reference Huston, Cutmore, Miller, Sasal, Smit and Cribb2021). Although such a level of intraspecific 28S variability is high, it is not unprecedented; Huston et al (Reference Huston, Cutmore, Cribb, Sasal and Yong2025) interpreted differences of 1–10 bp (0.08–0.79%) as intraspecific for specimens of Emprostiotrema fusum (Goto & Ozaki, 1929) Cianferoni & Ceccolini, 2021 (Emprostiotrematidae), though the geographic range in that case spanned across the Pacific. Considering all available data, and following the paradigm of Bray et al (Reference Bray, Cutmore and Cribb2022), we think these specimens are best considered conspecific. The criteria of Bray et al (Reference Bray, Cutmore and Cribb2022) require reciprocal monophyly in the most informative molecular marker plus either morphological distinction or biological distinction (in the form of differing host distributions). Despite specimens of N. pomacanthi forming four distinct lineages in the cox1 analysis, all these lineages form a single monophyletic clade in all analyses, are from the same hosts, and have the same morphology; thus, they do not meet the criteria for different species. Comparable interpretations have been made for other trematode groups in a range of recent studies in the tropical Indo-West Pacific (e.g., Bray et al Reference Bray, Cutmore and Cribb2023; Cribb et al Reference Cribb, Bray, Justine, Reimer, Sasal, Shirakashi and Cutmore2022; Cutmore et al Reference Cutmore, Corner and Cribb2023; Duong et al Reference Duong, Cribb and Cutmore2023; Huston et al Reference Huston, Cutmore and Cribb2022, Reference Huston, Cutmore, Cribb, Sasal and Yong2025).

Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) identified the specimens from Pomacanthus semicirculatus from Japan reported by Machida and Uchida (Reference Machida and Uchida2001) to be N. pomacanthi, rather than N. tsushimaensis as originally reported. This was based on the body size, length of the oesophagus and host identity. While we accept their interpretation, we draw attention to the strikingly different body shape of the Japanese worms (elongate fusiform vs pyriform). The specimens reported by Hafeezullah and Siddiqi (Reference Hafeezullah and Siddiqi1970) and Cribb et al. (Reference Cribb, Anderson and Bray1999, Reference Cribb, Bray, Hall and Cutmore2015) all have a distinct elongate fusiform shape, with the widest point in the posterior half of the body, at the level of the ventral sucker. The specimens of Machida and Uchida (Reference Machida and Uchida2001) are distinctly pyriform and narrow anteriorly from the level of the vitelline follicles. This shape could be an artifact of flattening, as Machida and Uchida (Reference Machida and Uchida2001) noted their samples were fixed under slight pressure. Ultimately, unflattened or genetically characterised specimens from Okinawa are required to ascertain if the range of N. pomacanthi does extend to Japanese waters.

Neoparantorchis tsushimaensis (Machida, 1971) n. comb.

Synonyms: Parantorchis tsushimaensis Machida, 1971; Antorchis tsushimaensis (Machida, 1971) Machida, 1975.

Type-host: Chaetodontoplus septentrionalis (Temminck & Schlegel), Bluestriped angelfish (Acanthuriformes: Pomacanthidae).

Other host: Centropyge flavissima (Cuvier), Lemonpeel angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Tsushima Island, Japan.

Other locality: New Caledonia.

Remarks

Machida (Reference Machida1971) described this species from Chaetodontoplus septentrionalis from Tsushima Island, Japan, and subsequently reported it from the same host and locality a few years later (Machida Reference Machida1975). In both reports it is noted that the caeca end anterior to, or at the level of the testes, that the ovary is post-testicular and, although not noted specifically in the descriptions, it can be elucidated from the illustrations that these worms have many vitelline follicles per side [at least 20 per side for the specimen illustrated by Machida (Reference Machida1971), and at least 15 per side for the specimen illustrated by Machida (Reference Machida1975)]. These features suggest that this species conforms to the concept of Neoparantorchis. Machida and Uchida (Reference Machida and Uchida2001) reported this species from the threespot angelfish, Apolemichthys trimaculatus (Cuvier), and Pomacanthus semicirculatus, from Okinawa, Japan. As noted above (see Remarks for N. pomacanthi), Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) re-examined the material from Japan and identified the specimens from Pomacanthus semicirculatus as Neoparantorchis pomacanthi, rather than N. tsushimaensis; we agree with this designation. In addition, based on the illustration provided by Machida and Uchida (Reference Machida and Uchida2001), we think that the specimens from Apolemichthys trimaculatus also do not represent N. tsushimaensis. Samples from A. trimaculatus have caeca that extend posteriorly past the testes, an ovary that is anteriorly intertesticular, and few, large vitelline follicles (9–12 per side); together these features clearly align with Parantorchis rather than Neoparantorchis. Given that the samples reported by Machida and Uchida (Reference Machida and Uchida2001) were flattened and are not represented by molecular data, whether these samples relate to one of the four species of Parantorchis or a new form is currently unknowable. We also re-examined the material from the lemonpeel angelfish, Centropyge flavissima (Cuvier), from New Caledonia reported by Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) and can provide no convincing reason to consider the specimens distinct from those representing N. tsushimaensis. We do, however, agree with Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) that ultimately the specimens from Chaetodontoplus septentrionalis and Centropyge flavissima may represent different species, but that such a split would require molecular confirmation or new, unflattened specimens.

Nasoantorchis n. g.

Diagnosis

Body elongate, oval. Tegument spinose. Oral sucker subglobular, subterminal. Ventral sucker not strongly muscular, in middle or posterior third of body. Prepharynx short. Pharynx oval, small. Oesophagus short, slender. Intestinal bifurcation in mid-forebody. Caeca short, saccular, thick-walled, terminate anterior to testes or at level of anterior margin of testes, never extend posteriorly beyond middle of testes. Testes subglobular, opposite, in forebody or at level of ventral sucker. Cirrus-sac entirely in forebody or extends to level of ventral sucker, distinctly recurved to form inverted U-shape. Internal seminal vesicle bipartite. Genital pore median, midway between ventral sucker and intestinal bifurcation. Ovary subglobular, post-testicular or intratesticular at the level of the posterior testicular margin, submedial. Laurer’s canal present, opens dorsally. Uterine coils extensive posterior to caeca. Eggs tanned, operculate. Vitelline follicles in two lateral clusters in forebody, each cluster comprising a large number (>25) of small follicles. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of acanthurid fishes.

Type and only species: Nasoantorchis nasonis (Cribb, Bray, Hall & Cutmore, 2015) n. comb.

ZooBank LSID: urn:lsid:zoobank.org:act:0659F005-4788-4D45-957E-47DBB73FE62A.

Etymology: This genus is named for the host genus, Naso Lacépède, and its morphological similarity to Antorchis. It should be treated as masculine.

Remarks

Although 28S data could not be generated for the sole species of this genus, analyses of cox1 data demonstrate that Nasoantorchis nasonis is only distantly related to species of Neoparantorchis and Parantorchis. Based on this molecular distinction, and differences in morphology (see Discussion) and host identity (acanthurids vs pomacanthids and chaetodontids), we think it best to consider N. nasonis distinct from Antorchis, Neoparantorchis and Parantorchis.

Nasoantorchis is morphologically most similar to the concept of Antorchis sensu stricto. Species of both Nasoantorchis and Antorchis have the ventral sucker in the middle of the body, a recurved cirrus-sac that forms an inverted U-shape, short caeca, a post-testicular ovary, and vitelline follicle clusters that are comprised of a large number of small follicles. Nasoantorchis nasonis differs from the concept of Antorchis in body shape (oval vs fusiform), the shape and position of the oral sucker (subglobular and subterminal vs infundibuliform or oval-pyriform and terminal), the length of the oesophagus (short vs long) and the position of the genital pore (midway between ventral sucker and intestinal bifurcation vs immediately anterior to ventral sucker).

Nasoantorchis nasonis (Cribb, Bray, Hall & Cutmore, 2015) n. comb.

Synonyms: Antorchis nasonis Cribb, Bray, Hall & Cutmore, 2015.

Type-host: Naso annulatus (Quoy & Gaimard), Whitemargin unicornfish (Acanthuriformes: Acanthuridae).

Other host: Naso tonganus (Valenciennes), Bulbnose unicornfish (Acanthuriformes: Acanthuridae).

Type-locality: Off Heron Island, southern Great Barrier Reef, Australia.

Material lodged: 17 voucher specimens (G242492–508) lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518166.

Remarks

Nasoantorchis nasonis was described by Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) based on samples from two species of Naso from the Great Barrier Reef. Cribb et al (Reference Cribb, Bray, Hall and Cutmore2015) considered samples identified as Antorchis urna by Machida and Uchida (Reference Machida and Uchida1990), from the sleek unicornfish, Naso hexacanthus (Bleeker) from Japan, as highly similar to N. nasonis, but did not consider them conspecific based on slight morphological differences. If the form from Japanese waters is distinct from N. nasonis, we suspect it will also belong to Nasoantorchis given it has a recurved cirrus-sac, short caeca, many small vitelline follicles, and uterine coils that are extensive posterior to the caeca. Further collections of this species for the generation of 28S data would be highly beneficial in positioning it within the family.

Findovitellus n. g.

Diagnosis

Body elongate, fusiform. Tegument spinose, spines small. Oral sucker subglobular, subterminal. Ventral sucker large, distinctly squared, without anterior or posterior semicircular muscular appendage, in mid-body. Prepharynx short. Pharynx subglobular, small. Oesophagus moderately long. Intestinal bifurcation in mid-forebody. Caeca short, saccular, terminate in forebody or at level of anterior margin of ventral sucker. Testes subglobular, opposite, at level of posterior margin of ventral sucker. Cirrus-sac mostly in forebody, reaching just posterior to anterior margin of ventral sucker. Internal seminal vesicle bipartite. Pars prostatica narrow. Ejaculatory duct short. Genital atrium distinct. Genital pore median, near level of intestinal bifurcation. Ovary subglobular, pretesticular. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally. Uterine coils extensive in hindbody, not at all developed in forebody. Eggs tanned, operculate. Vitelline follicles distributed laterally in forebody, each lateral cluster distributed in anterior and posterior groups. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of pomacanthid fishes.

Type and only species: Findovitellus leichhardti (Diaz & Cribb, 2013) n. comb.

ZooBank LSID: urn:lsid:zoobank.org:act:26A8B4AC-3098-4F3F-9850-A929186F0843.

Etymology: This genus is named for the Latin findere (to separate) and vitellus (yolk), in reference to the lateral vitelline follicles being split into two groups. It should be treated as masculine.

Remarks

The proposal of Findovitellus, and Crassitrema (proposed below), both as distinct from Paradiscogaster, is fundamentally based on phylogenetic topology. In both the cox1 and 28S analyses, Paradiscogaster leichhardti and Paradiscogaster quasimodo resolve in a strongly supported clade distant from all other species of Paradiscogaster for which there are molecular data. In the 28S analyses, these two species form a clade with the highly morphologically distinct Trigonocryptus conus Martin, 1958; these three species are then sister to the clade of Neoparantorchis and Parantorchis species. Despite being highly distinct genetically, Paradiscogaster leichhardti and Paradiscogaster quasimodo both broadly agree with the current morphological diagnosis of Paradiscogaster. However, since it was proposed by Yamaguti (Reference Yamaguti1934), the diagnosis of Paradiscogaster has broadened significantly to accommodate the range of species added to the genus. Both P. leichhardti and P. quasimodo have morphological features that allow a distinction from all other known Paradiscogaster species and from each other, and we think it is appropriate to propose a new generic placement for both these species.

Paradiscogaster leichhardti and P. quasimodo do not agree with the diagnosis of the closest taxa in the phylogenetic analyses, Trigonocryptus, nor do they agree with the diagnoses of any other antorchiine genus. Thus, it is clear that a new generic concept is required for these species; however, the choice between a single new genus for both species, or two new genera, is less clear. Paradiscogaster leichhardti and P. quasimodo are mutually highly morphologically distinct. Paradiscogaster leichhardti is elongate and pyriform, with a distinctly squared ventral sucker and vitelline follicles that are distributed in four lateral fields (two on each side) in the forebody. Paradiscogaster quasimodo is short and oval, with a subglobular ventral sucker and vitelline follicles in two lateral fields on either side of the ventral sucker. Despite these morphological differences, the two species are each other’s closest relatives in all analyses and differ at just 29 bp (p-distance = 2.25%) in the 28S dataset. This level of difference among antorchiines is otherwise within the range of interspecific differences within a genus, rather than that between genera. Thus, relative to other genera included in the current analyses, the difference between P. leichhardti and P. quasimodo is less than those between the two species of Discogasteroides (38–44 bp), two species of Scaticola (76 bp), and several combinations of the 10 species of Paradiscogaster that do form a well-supported clade (up to 134 bp). Thus, it is arguable that the two represent congeneric species that have recently undergone dramatic morphological divergence. Proposing a shared genus for these two species, however, would result in a diagnosis that is entirely cryptic relative to the diagnosis of Paradiscogaster. We consider the diagnosability of the genera within the Antorchiinae important and therefore propose new genera to accommodate P. leichhardti and P. quasimodo, Findovitellus and Crassitrema, which can be morphologically distinguished.

Findovitellus leichhardti can be distinguished from all other antorchiine genera by the distribution of the vitelline follicles, which are distributed in two distinct groups: an anterior field at the level of the intestinal bifurcation and a posterior field at the level of the anterior margin of the ventral sucker. No other antorchiine genus has species in which the vitelline fields are divided into anterior and posterior fields. In addition, in the possession of an elongate, fusiform body, F. leichhardti differs from Parayamagutia and Trigonocryptus. In the possession of a large, distinctly squared ventral sucker, F. leichhardti differs from Antorchis, Nasoantorchis, Neoparantorchis, Parantorchis and Scaticola.

Findovitellus leichhardti (Diaz & Cribb, 2013) n. comb.

Type-host: Chaetodontoplus meredithi Kuiter, Queensland yellowtail angelfish (Perciformes: Pomacanthidae).

Type-locality: Off Heron Island, Great Barrier Reef, Australia.

Synonym: Paradiscogaster leichhardti Diaz & Cribb, 2013.

New material

Known host: Chaetodontoplus meredithi.

Known locality: Off Heron Island, southern Great Barrier Reef, Queensland, Australia (23°27′S, 151°55′E).

Site of infection: Intestine.

Prevalence: 2/10 (20%).

Deposition of specimens: Two voucher specimens (G242509–10), including one hologenophore, lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518200–02; ITS2 rDNA, PZ510936; 28S rDNA, PZ513246.

Remarks

This is just the second report of Findovitellus leichhardti, collected from the type-host at the type-locality. We note a few minor differences from the original description by Diaz and Cribb (Reference Diaz and Cribb2013). In the original description, the tegumental spines are described as only reaching to the anterior forebody; in the new specimens, the spines are present in the entire hindbody but become progressively sparser from the ventral sucker to the posterior end of the body. The caeca were described as terminating in the forebody; in the new specimens, they can reach just posterior to the level of the anterior margin of the ventral sucker. The cirrus-sac was described as being entirely in the forebody; we found that it could also extend to just posterior to the anterior margin of the ventral sucker. The ovary was described as pretesticular; we found it to be pretesticular or intertesticular.

Crassitrema n. g.

Diagnosis

Body short, oval. Tegument smooth, spines absent. Oral sucker subglobular, subterminal. Ventral sucker large, close to overlapping oral sucker, in form of indented muscular pad with lobate margins. Forebody short, almost absent. Prepharynx short. Pharynx small, subglobular. Oesophagus recurved dorsally. Intestinal bifurcation dorsal to ventral sucker. Caeca dorsal to ventral sucker. Testes subglobular, opposite, at level of posterior margin of ventral sucker. Cirrus-sac dorsal to ventral sucker. Internal seminal vesicle bipartite. Pars prostatica tubular. Ejaculatory duct long. Genital atrium distinct. Genital pore median, at anterior margin of ventral sucker. Ovary subglobular, intertesticular at level of posterior testicular margin. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally on distinct nipple-like muscular protuberance. Uterine coils extensive in hindbody, not at all developed in forebody. Eggs tanned, operculate. Vitelline follicles distributed in lateral clusters, at level of ventral sucker. Excretory vesicle Y-shaped. Excretory pore terminal. In intestine of chaetodontid fishes.

Type and only species: Crassitrema quasimodo (Bray, Cribb & Barker, 1994) n. comb.

ZooBank LSID: urn:lsid:zoobank.org:act:B2C47171-116E-492D-881D-97D2C54397D4.

Etymology: This genus is named for the Latin adjective crassus (fat or plump), in reference to the compressed morphology of the sole species of this genus, and the Latin neuter noun trema (an opening, hole, cave) commonly used in trematode generic names. The generic name is neuter.

Remarks

The basis for the proposal of Crassitrema is discussed above in the Remarks section for Findovitellus. In brief, despite limited genetic distinction between Paradiscogaster leichhardti and P. quasimodo, we argue that the morphological differences are too great for them to be considered congeners. The type and only species of this new genus, Crassitrema quasimodo, can be distinguished from almost all other antorchiine by its body shape. Crassitrema quasimodo is compressed antero-posteriorly, making this species appear plump and oval in both dorsoventral and lateral views. The ventral sucker is large and anteriorly oriented, which results in virtually no forebody. This body shape differentiates it from species of Discogastroides, Findovitellus, Nasoantorchis, Neoparantorchis, Paradiscogaster, Parantorchis and Scaticola. Species of Trigonocryptus also have a compressed body shape but have a distinctly triangular, conical body with an enormous ventral sucker that pulls the lateral margins of the worm ventrally to overlap a significant portion of the body. In terms of the possession of a plump body that is oval, it is similar to the type and only species of Parayamagutia, P. ostracionis; however, it differs from P. ostracionis in having almost no forebody (vs a long forebody), a uterus that is restricted to the hindbody (vs one that extends to the level of the oral sucker) and vitelline follicles at the level of the ventral sucker (vs at the level of the oral sucker).

Paradiscogaster Yamaguti, 1934

Diagnosis

See emended diagnosis in Bray (Reference Bray, Bray, Gibson and Jones2008).

Type-species: Paradiscogaster pyriformis Yamaguti, 1934.

Other species: Paradiscogaster aluteri Machida, 1972; Paradiscogaster arabi (Hafeezullah and Siddiqi, 1970) Cribb, Anderson & Bray, 1999; Paradiscogaster brayi Diaz, Cutmore & Cribb, 2017; Paradiscogaster chaetodontis Yamaguti, 1938; Paradiscogaster drepanei Mamaev, 1970; Paradiscogaster dweorg Cribb, Anderson & Bray, 1999; Paradiscogaster eniwetokensis Martin & Hammerich, 1970; Paradiscogaster farooqii Hafeezullah & Siddiqi, 1970; Paradiscogaster flindersi Bray, Cribb & Barker, 1994; Paradiscogaster glebulae Bray, Cribb & Barker, 1994; Paradiscogaster habilis Cribb, Anderson & Bray, 1999; Paradiscogaster hainanensis (Shen, 1990) Cribb, Anderson & Bray, 1999; Paradiscogaster indicus (Srivastava, 1939) Cribb, Anderson & Bray, 1999; Paradiscogaster lobomyzon Cribb, Anderson & Bray, 1999; Paradiscogaster machidai Cribb, Anderson & Bray, 1999; Paradiscogaster mannari Ramesh, Saravanan, Nandiath & Madhavi, 2019; Paradiscogaster manteri Kurochkin, 1970; Paradiscogaster martini Diaz, Cutmore & Cribb, 2017; Paradiscogaster melendezi Diaz, Bray, Cutmore, Ward & Cribb, 2015; Paradiscogaster munozae Diaz, Bray, Cutmore, Ward & Cribb, 2015; Paradiscogaster nitschkei Diaz, Cutmore & Cribb, 2017; Paradiscogaster oxleyi Diaz, Bray & Cribb, 2013; Paradiscogaster pritchardae (Gupta & Ahmad, 1978) Cribb, Anderson & Bray, 1999; Paradiscogaster vichovae Diaz, Cutmore & Cribb, 2017; Paradiscogaster yamagutii Hafeezullah & Siddiqi, 1970.

Species inquirenda: Paradiscogaster niger Bilqees, 1976.

Remarks

Paradiscogaster was originally proposed by Yamaguti (Reference Yamaguti1934) for P. pyriformis Yamaguti, 1934, collected from a pleuronectid, the ridged-eye flounder, Pleuronichthys cornutus (Temminck & Schlegel), from Japan. Since then, species from 12 further fish families (Carangidae, Chaetodontidae, Drepaneidae, Ephippidae, Kyphosidae, Monacanthidae, Monodactylidae, Ostraciidae, Pomacanthidae, Scatophagidae, Terapontidae, and Triacanthidae) have been assigned to this genus. Prior to this study, Paradiscogaster comprised 35 species, and it is, by far, the richest genus in the Antorchiinae. Whilst we have proposed a new generic status for five of the known species, it seems unlikely that all of the remaining species in Paradiscogaster will prove to be congeneric; the varied morphology shown by some Paradiscogaster species, and the wide range of unrelated hosts these species infect, is certainly atypical of what is normally understood for congeneric trematodes in the tropical Indo-West Pacific. Most important in clarifying the status of these species is the recollection and sequencing of the type-species, given that it is the only Paradiscogaster species known from a pleuronectid. This species has been convincingly re-collected just once since its description; Yamaguti (Reference Yamaguti1951) collected eight specimens from the type-host close to the type-locality, off Tokoname (Aichi Prefecture), Japan.

Paradiscogaster machidai Cribb, Anderson & Bray, 1999 (Figure 5a)

Figure 5.

Species of Paradiscogaster Yamaguti, 1934 ex pomacanthids from the northern Great Barrier Reef and Ningaloo Reef, Australia. (a) Paradiscogaster machidai Cribb, Anderson & Bray, 1999 ex Pomacanthus semicirculatus (Cuvier) from off Lizard Island, ventral view; (b) Paradiscogaster hoggettae n. sp. ex Pomacanthus imperator (Bloch) from Yonge Reef, ventral view; (c) Paradiscogaster ningalooensis n. sp. ex Pomacanthus imperator (Bloch) from Ningaloo Reef, ventral view. Scale bars = 500 μm.

Figure 5. Long description

A multi-panel figure containing three detailed line drawings of parasitic flatworms. Each drawing includes a vertical scale bar representing 500 micrometers.

* Panel a, top left, shows Paradiscogaster machidai. It has an elongated, pear-shaped body. At the anterior end is a large, circular oral sucker with radial striations. Moving posteriorly, a short esophagus leads to a muscular pharynx and a bifurcated intestine. Dark, clustered vitelline follicles are concentrated along the lateral margins in the mid-body. The posterior half is filled with a dense mass of small, oval eggs.

* Panel b, bottom center, shows Paradiscogaster hoggettae. This species has a broader, more diamond-shaped body compared to panel a. The oral sucker is similarly positioned at the anterior tip. The vitelline follicles are more widely dispersed across the lateral fields. The central region shows a coiled uterus filled with eggs, and a distinct ventral sucker is visible in the mid-body region.

* Panel c, top right, shows Paradiscogaster ningalooensis. The body shape is intermediate between a and b. The internal arrangement is similar, with the oral sucker at the top, followed by the digestive and reproductive systems. The vitelline follicles are clustered in the lateral fields, and the posterior region is packed with numerous eggs.

All three specimens exhibit a small notch at the posterior-most tip representing the excretory pore.

Type-host: Pomacanthus semicirculatus (Cuvier), Blue angelfish (Acanthuriformes: Pomacanthidae).

Other hosts: Pomacanthus sexstriatus (Cuvier), Sixbar angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Off Heron Island (23°27′S, 151°55′E), southern Great Barrier Reef, Australia.

New material

New host: Centropyge tibicen (Cuvier), Keyhole angelfish (Acanthuriformes: Pomacanthidae).

Known hosts: Pomacanthus semicirculatus; Pomacanthus sexstriatus.

New localities: Off Lizard Island (14°40′S, 145°27′E), northern Great Barrier Reef, Queensland, Australia. Ningaloo Reef (23°12′S 113°45′E), Western Australia.

Known locality: Off Heron Island.

Site of infection: Intestine.

Prevalence: Heron Island: Centropyge tibicen, 1/2 (50%). Lizard Island: Pomacanthus semicirculatus, 1/1 (100%); Pomacanthus sexstriatus, 1/1 (100%). Ningaloo Reef: Pomacanthus sexstriatus, 1/2 (50%).

Deposition of specimens: 24 voucher specimens (G242511–34) lodged in the QM; four voucher specimens (V13678–81) lodged in the WAM.

Molecular sequence data: cox1 mtDNA, PZ518190–94; ITS2 rDNA, PZ510930–32; 28S rDNA, PZ513259–60.

Remarks

Specimens of Paradiscogaster machidai were collected from two known hosts (Pomacanthus semicirculatus and Pomacanthus sexstriatus) and one new host (Centropyge tibicen) from off Heron Island, Lizard Island and Ningaloo Reef. The collections from off Lizard Island and Ningaloo Reef expand the known range of this species. cox1 sequence data were generated for all host/locality combinations; these data formed a well-supported clade with intraclade variation at 0–5 bp (p-distance = 0–1.06%) and do not resolve by geographical location. The samples from the Great Barrier Reef and Ningaloo Reef differ, however, at 2 bp in the ITS2 dataset and 3 bp in the 28S dataset. This is the first report of the species since its original description.

Paradiscogaster hoggettae n. sp. (Figure 5b)

Type-host: Pomacanthus imperator (Bloch), Emperor angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Yonge Reef (14°35′S, 145°37′E), northern Great Barrier Reef, Queensland, Australia.

Site of infection: Intestine.

Prevalence: 1/1 (100%).

Deposition of specimens: Holotype (G242535) and seven paratypes (G242536–42), including hologenophores, lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518195; ITS2 rDNA, PZ510933; 28S rDNA, PZ513262.

ZooBank LSID: urn:lsid:zoobank.org:act:B677DC31-A82D-4490-A8C9-0D8D8E25039E.

Etymology: This species is named for Dr Anne Hoggett, who, with her husband, Dr Lyle Vail, is the recently retired joint Director of the Lizard Island Research Station. We name this species for Anne in appreciation of a lifetime’s contribution to fostering, facilitating and promoting Australian marine science.

Description

[Measurements based on five wholemounts and three hologenophore partial mounts.] Body pyriform, widest at level of ventral sucker, 1,032–1,253 × 584–782 (1,162 × 697); body length to width ratio 1.68–1.77 (1.71):1. Tegument spinose; spines smaller in hindbody. Forebody 464–637 (563) long, occupying 45–48 (46)% of body length. Oral sucker subspherical, 142–200 × 177–223 (176 × 206). Ventral sucker a transversely oval muscular pad, usually without evident cavity, mainly in posterior half of body, 325–402 × 555–739 (377 × 651), spans 90–97 (94)% of body width; oral sucker length to ventral sucker length ratio 0.44–0.54 (0.47):1; oral sucker width to ventral sucker width ratio 0.30–0.33 (0.32):1. Prepharynx short, always within posterior cavity of oral sucker. Pharynx small, oval, 62–72 × 86–101 (67 × 94); pharynx length to oral sucker length ratio 0.35–0.44 (0.38):1; pharynx width to oral sucker width ratio 0.41–0.50 (0.46):1. Oesophagus short, straight or slightly sinuous, 104–162 (119) long. Intestinal bifurcation in mid-forebody. Caeca medium length, saccular, overlap anterior half of ventral sucker; left caecum 351–465 (427) long; right caecum 362–466 (431) long; caecal field 300–412 (379) long, occupying 29–34 (32)% of body length; caecal length to oesophagus length 2.58–4.43 (3.71):1.

Testes two, round to ovoid, entire, opposite, at level of posterior margin of ventral sucker; left testis 72–147 × 72–137 (112 × 102); right testis 65–135 × 65–149 (106 × 95); post-testicular space 190–237 (218) long, or 15–20 (19)% of body length. Cirrus-sac oval, medial, extends from level of anterior third of ventral sucker to common genital pore at about level of intestinal bifurcation, 284–425 × 153–231 (343 × 193). Internal seminal vesicle bipartite; anterior portion larger, saccular, 41–110 × 50–142 (63 × 77); posterior portion tubular. Pars prostatica straight, prominent. Eversible ejaculatory duct short. Genital atrium distinct. Genital pore medial or just submedial, immediately anterior to level of intestinal bifurcation, 265–354 (304) from anterior margin of body, 770–962 (888) from posterior margin of body.

Ovary dorsal to posterior half of ventral sucker, medial to submedial, entire, between testes, 44–139 × 47–136 (98 × 92); post-ovarian space 251–326 (279) long. Canalicular seminal receptacle present, medial to submedial, dorsal to ventral sucker. Laurer’s canal opens dorsally close to posterior end of body. Dorsal invagination prominent, just posterior to opening of Laurer’s canal. Vitelline follicles in two separate lateral fields, globular, 19–28 (23) in diameter; vitelline fields extend from near level of anterior margin of ventral sucker to just anterior to genital pore, 232–406 (324) long, occupying 23–29 (26)% of body length, 314–413 (352) from anterior extremity, 439–564 (518) from posterior extremity. Uterus extensive, fills most space posterior to middle of ventral sucker; final loop of uterus runs medially, parallel to cirrus-sac. Eggs tanned, operculate, 26–33 × 15–18 (29 × 17).

Excretory pore terminal. Excretory vesicle Y-shaped; stem short; arms terminate close to posterior margin of testes.

Remarks

Specimens of Paradiscogaster hoggettae were collected from just a single host and locality, Pomacanthus imperator from Yonge Reef; no specimens of this species were found in two P. imperator specimens from Ningaloo Reef or one from French Polynesia. There was no variation in the cox1, ITS2, and 28S data for specimens of this species.

In the possession of a distinctly pyriform body and a ventral sucker that is a very large, transversely oval muscular pad in the posterior half of the body, P. hoggettae resembles P. chaetodontis, P. drepanei, P. dweorg, P. flindersi, P. hainanensis, P. machidai, P. oxleyi, and P. pyriformis. The ventral sucker of P. hoggettae is massive, spanning almost the entire body width ventrally, differentiating it from P. chaetodontis, P. drepanei, P. flindersi, P. hainanensis, P. oxleyi, and P. pyriformis. Paradiscogaster dweorg and P. machidai are the only other species with a ventral sucker as extensive as the new species; Paradiscogaster hoggettae differs from P. dweorg in forebody length and shape (long and distinctly narrowing vs short and broad), caecal shape and length (moderate length and narrow vs short and wide) and the position of the ventral sucker (in posterior half of body vs anteriorly reaching to level of the pharynx).

Paradiscogaster machidai, described from Pomacanthus semicirculatus and Pomacanthus sexstriatus from the Great Barrier Reef, is the species that is genetically and ecologically most similar to P. hoggettae. Based on morphometrics alone, the species are almost mutually cryptic. Paradiscogaster machidai, P. hoggettae and another new species described below, P. ningalooensis, represent a complex of species that require a detailed evaluation of morphometrics, molecular data and host identity to delineate. Delineation of these species is thus partly subjective, requiring interpretation beyond the criterion proposed by Bray et al. (Reference Bray, Cutmore and Cribb2022), which is provided in the remarks section for Paradiscogaster ningalooensis (see below).

Paradiscogaster ningalooensis n. sp. (Figure 5c)

Type-host: Pomacanthus semicirculatus (Cuvier), Blue angelfish (Acanthuriformes: Pomacanthidae).

Other host: Pomacanthus imperator (Bloch), Emperor angelfish (Acanthuriformes: Pomacanthidae).

Type-locality: Ningaloo Reef (23°12′S, 113°45′E), Western Australia.

Site of infection: Intestine.

Prevalence: P. semicirculatus, 2/3 (66%); P. imperator, 2/2 (100%).

Deposition of specimens: Holotype (13682), nine paratypes (13683–91) and 14 vouchers (V13692–705), including hologenophores, lodged in the WAM.

Molecular sequence data: cox1 mtDNA, PZ518196–99; ITS2 rDNA, PZ510934–35; 28S rDNA, PZ513261.

ZooBank LSID: urn:lsid:zoobank.org:act:F619A063-AA96-4E25-A8B7-9570D1BAE00C.

Etymology: This species is named for its type-locality, Ningaloo Reef.

Description

[Measurements based on 10 wholemounts.] Body pyriform, widest at level of ventral sucker, 1,025–1,488 × 463–744 (1,256 × 593); body length to width ratio 1.94–2.31 (2.13):1. Tegument spinose; spines smaller in hindbody. Forebody 463–744 (578) long, occupying 43–50 (46)% of body length. Oral sucker subspherical, 138–213 × 169–266 (163 × 200). Ventral sucker a transversely oval muscular pad, usually with only shallow cavity, mainly in posterior half of body, 350–475 × 463–725 (410 × 583), spans 93–100 (98)% of body width; oral sucker length to ventral sucker length ratio 0.30–0.50 (0.40):1; oral sucker width to ventral sucker width ratio 0.30–0.38 (0.34):1. Prepharynx short, usually within posterior cavity of oral sucker. Pharynx small, oval, 50–88 × 56–97 (72 × 81); pharynx length to oral sucker length ratio 0.32–0.55 (0.45):1; pharynx width to oral sucker width ratio 0.29–0.50 (0.41):1. Oesophagus short, straight, 63–125 (85) long. Intestinal bifurcation in mid-forebody. Caeca medium length, saccular, overlap anterior half of ventral sucker; left caecum 288–500 (31) long; right caecum 288–469 (375) long; caecal field 275–466 (364) long, occupying 27–34 (30)% of body length; caecal length to oesophagus length 3.15–7.24 (4.65):1.

Testes two, round to ovoid, entire, opposite, at level of posterior margin of ventral sucker; left testis 113–194 × 103–153 (142 × 114); right testis 119–181 × 100–156 (145 × 123); post-testicular space 197–306 (243) long, or 17–22 (19)% of body length. Cirrus-sac oval, medial, extends from level of anterior third of ventral sucker to common genital pore at about level of intestinal bifurcation, 250–447 × 147–275 (350 × 192). Internal seminal vesicle bipartite; anterior portion larger, saccular, 63–94 × 63–125 (71 × 84); posterior portion tubular. Pars prostatica straight, prominent. Eversible ejaculatory duct short. Genital atrium distinct. Genital pore medial or just submedial, immediately anterior to level of intestinal bifurcation, 225–394 (306) from anterior margin of body, 756–1,122 (925) from posterior margin of body.

Ovary dorsal to posterior half of ventral sucker, medial to submedial, entire, between testes, 69–125 × 81–131 (99 × 110); post-ovarian space 300–450 (360) long. Canalicular seminal receptacle present, medial to submedial, dorsal to ventral sucker. Laurer’s canal opens dorsally close to posterior end of body. Dorsal invagination prominent, just posterior to opening of Laurer’s canal. Vitelline follicles in two separate lateral fields, globular, 15–33 (24) in diameter; vitelline fields extend from level of anterior half of ventral sucker to just anterior to genital pore, 325–563 (437) long, occupying 26–40 (35)% of body length; 175–344 (260) from anterior extremity; 438–694 (544) from posterior extremity. Uterus extensive, fills most space posterior to middle of ventral sucker; final loop of uterus runs medially, parallel to cirrus-sac. Eggs tanned, operculate, 23–28 × 13–15 (25 × 14).

Excretory pore terminal. Excretory vesicle Y-shaped; stem short; arms terminate close to posterior margin of testes.

Remarks

Specimens of Paradiscogaster ningalooensis were collected from Pomacanthus imperator and P. semicirculatus from just a single locality, Ningaloo Reef; no specimens of this species were found in either host examined from Lizard Island or French Polynesia. There was minor intraspecific variation in the cox1 data for this species, with samples differing at 0–2 bp (p-distance = 0–0.42%); these differences do not relate to host identity. There are no differences in ITS2 data, and just one 28S sequence was generated for this species.

Morphologically, Paradiscogaster ningalooensis is almost cryptic relative to P. hoggettae and P. machidai; it is so similar that this new species can be delineated from all other Paradiscogaster species (except for P. hoggettae and P. machidai) using the delineation for P. hoggettae provided above. These three species represent a problematic complex for the identification of species boundaries, which is confused by host identity, locality, morphology and phylogenetic topology. The three species are highly ecologically similar, all infecting species of Pomacanthus, but do not have entirely overlapping host ranges; P. hoggettae is known only from Pomacanthus imperator, P. ningalooensis is known only from Pomacanthus imperator and P. semicirculatus, and P. machidai is known only from Pomacanthus semicirculatus, Pomacanthus sexstriatus, and Centropyge tibicen. Given our low collection numbers, especially for P. imperator, it is impossible to predict whether these distinctions have any meaning. All three species occur in Australian waters, but the three are not entirely sympatric; P. ningalooensis and P. machidai occur together at Ningaloo Reef, and P. hoggettae and P. machidai occur together on the northern Great Barrier Reef; P. ningalooensis and P. hoggettae have not been found in sympatry. In all phylogenetic analyses, the three species resolve as each other’s closest relatives, and in the cox1 dataset the clades of the three species differ at fewer positions than between what is interpreted as the conspecific clades of Neoparantorchis pomacanthi (collected from the same pomacanthid hosts from the same localities). Despite these confounding issues, there are some morphometric and phylogenetic attributes that suggest that three species should be recognised.

First, despite being nearly morphometrically cryptic, there is a combination of subtle morphological characters that distinguish the three species. Each of these characters is shared between some combination of the three species, but considered together they suggest that each species has its own distinct morphology. First, although the ventral sucker is large for P. machidai, it is larger for P. ningalooensis and P. hoggettae (Figure 6). For the two new species, the ventral sucker spans almost the entire ventral body width (mean % of body width = 98% and 94%, respectively); in contrast, there is a distinct space between the lateral margins of the ventral sucker and the lateral margins of the body for P. machidai. The body length to width ratio is greater for P. machidai (mean = 2.81:1) and P. ningalooensis (mean = 2.13:1) than for P. hoggettae (mean = 1.71:1); as a result, P. hoggettae appears much more pyriform than P. machidai and P. ningalooensis (Figure 6). Paradiscogaster hoggettae has a longer oesophagus than P. ningalooensis and P. machidai (mean length = 119 vs 85 and 87, respectively). Paradiscogaster ningalooensis has a longer vitelline field than P. hoggettae (mean = occupying 35% vs 26%) and a longer post-ovarian space (mean length = 360 vs 279). Finally, the eggs of Paradiscogaster ningalooensis are, on average, slightly smaller than those of P. hoggettae (mean = 25 × 14 vs 29 × 17), but the ranges for these species overlap, and both overlap with the eggs of P. machidai.

Figure 6.

Outlines of species of Paradiscogaster Yamaguti, 1934 ex pomacanthids, demonstrating differences in body shape and ventral sucker size. (a–d) Paradiscogaster machidai Cribb, Anderson & Bray, 1999; (e–h) Paradiscogaster hoggettae n. sp.; (i–l) Paradiscogaster ningalooensis n. sp. Scale bars = 500 μm.

Figure 6. Long description

A multi-panel line drawing containing 12 specimens labeled a through l.

Row 1 (a-d) Paradiscogaster machidai: These specimens have elongated, oval bodies. At the anterior end is a circular oral sucker. Below this, a short esophagus bifurcates into two intestinal caeca that extend down to a large, wide ventral sucker occupying the middle third of the body. The posterior end tapers to a small point.

Row 2 (e-h) Paradiscogaster hoggettae: These specimens exhibit a more pear-shaped or rhomboid body, wider at the level of the ventral sucker than the first row. The ventral sucker is notably larger and broader, filling nearly the entire width of the mid-body. The anterior region is more constricted.

Row 3 (i-l) Paradiscogaster ningalooensis: These specimens show a transition in shape from elongated (i, j) to more robust and rounded (k, l). The ventral sucker remains large and prominent in the lower half of the body.

Vertical scale bars representing 500 micrometers are positioned to the right of each row. All specimens share a common internal structure consisting of an oral sucker, a pharynx, a bifurcating digestive tract, and a large ventral sucker.

In terms of genetic distinction, the different gene regions provide contradictory data. The three species differ at 21–27 bp (p-distance = 4.43–5.70%) in the cox1 dataset, with phylogenetic analysis on these data resolving P. hoggettae and P. ningalooensis as sister clades; this clade is then sister to P. machidai. Such a level of cox1 difference between the three species is particularly low to be considered interspecific; 4.43–5.70% is less than some of the differences interpreted as intraspecific between clades of single species in this study (e.g., between the four clades Neoparantorchis pomacanthi) and for species of other families in the same region (e.g., Cutmore and Cribb Reference Cutmore and Cribb2022; Cutmore et al. Reference Cutmore, Corner and Cribb2023; Magro et al. Reference Magro, Cutmore, Carrassón and Cribb2023; Huston et al. Reference Huston, Cutmore, Cribb, Sasal and Yong2025). Notably, samples of Paradiscogaster machidai show almost no cox1 variation over range, with samples from the Great Barrier Reef and Ningaloo Reef (located on different sides of the Australian continent) differing at just 0–1.06%. The most informative molecular data are, arguably, the 28S sequences. The low differences between the three species in the cox1 dataset correspond to comparatively large differences in the 28S dataset; the three species differ at 14–22 bp (p-distance = 1.08–1.70%). Such a difference is much higher than is typically considered conspecific, even over large geographical ranges, and is similar to interspecific differences for a range of families in the tropical Indo-West Pacific (e.g., Cribb et al. Reference Cribb, Bray, Justine, Reimer, Sasal, Shirakashi and Cutmore2022; Cribb et al. Reference Cribb, Martin and Cutmore2025; Cribb et al. Reference Cribb, Martin, Diaz, Bray and Cutmore2021; Wee et al. Reference Wee, Cribb, Shirakashi and Cutmore2022). Conspicuously, the four clades of Neoparantorchis pomacanthi that had substantial distinction in the cox1 dataset differed at just 2–9 bp (p-distance = 0.16–0.71%) in the 28S dataset. Currently, we think the best interpretation of these samples is that they represent three species in an almost morphologically cryptic complex that are generally distinguishable by subtle morphology and ribosomal genetic data. Subtle morphological distinction combined with substantial 28S rDNA differences was decisive in the formation of this hypothesis.

Discogastroides Strand, 1934

Emended diagnosis

Body short, pyriform. Tegument spinose, spines small or massive. Oral sucker subglobular and subterminal or cup-shaped and terminal with distinct lip. Ventral sucker a large, muscular pad with concave anterior and posterior margins or projections on anterior and posterior margins, in posterior half of body. Prepharynx short. Pharynx subglobular, small. Oesophagus short. Intestinal bifurcation in anterior to mid-forebody. Caeca short, saccular, terminate at level of anterior margin of ventral sucker. Testes subglobular, opposite, in hindbody near level of posterior portion of ventral sucker. Cirrus-sac extends posteriorly to level of ventral sucker. Internal seminal vesicle bipartite. Pars prostatica long. Ejaculatory duct short. Genital atrium distinct. Genital pore median or sinistrally or dextrally submedial, at level of intestinal bifurcation. Ovary subglobular, at level of posterior half of ventral sucker. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally. Uterine coils extensive in hindbody. Eggs tanned, operculate. Vitelline follicles distributed in lateral fields; fields extend from mid-forebody to posterior margin of ventral sucker. Excretory vesicle saccate or Y-shaped. Excretory pore terminal. In intestine of ostraciid fishes.

Type-species: Discogastroides ostracii (Yamaguti, 1934) Skrjabin & Koval, 1957.

Other species: Discogastroides macrostomus Shimazu & Kamegai, 1990.

Remarks

Discogastroides was proposed by Strand (Reference Strand1934) as a replacement name for the homonym Discogaster Yamaguti, 1934 nec Burmeister, 1835, which was proposed the same year for Discogaster ostracii and Discogaster minor Yamaguti, 1934, both described from the roundbelly cowfish, Lactoria diaphana (Bloch & Schneider), from Japanese waters. The genus was expanded with a further six species described from fishes of the Carangidae, Chaetodontidae, Monacanthidae, Ostraciidae and Tetraodontidae. Cribb et al (Reference Cribb, Anderson and Bray1999) considered the morphological distinctions between Discogastroides and Paradiscogaster as species-level rather than generic-level differences and synonymised Discogastroides. In the new phylogenetic analyses, the three antorchiine zoogonids known from ostraciids form a highly supported clade; the morphologically distinct Parayamagutia ostracionis is sister to a clade of the other two species, D. ostracii and D. macrostomus.

Morphologically, Discogastroides and Paradiscogaster are highly similar. Yamaguti (Reference Yamaguti1934) delineated the two genera on the length of the oesophagus, the form of the cirrus-sac and the position of the vitellarium. While we agree with Cribb et al (Reference Cribb, Anderson and Bray1999) that the length of the oesophagus and the form of the cirrus-sac fail to distinguish these taxa at a generic level, the position of the vitellarium appears to be distinctive. As described and illustrated by Yamaguti (Reference Yamaguti1934), the vitelline follicles of the type-species of Discogastroides, Discogastroides ostracii, are positioned marginally at the level of the ventral sucker, while for the type-species of Paradiscogaster, Paradiscogaster pyriformis, they are positioned submarginally at the level of the caeca. The vitellarium of Discogastroides macrostomus resembles that of Discogastroides ostracii, being marginally restricted, and these two species form a well-supported clade in all analyses. Although Paradiscogaster pyriformis is not represented in the phylogenetic analysis, all species of Paradiscogaster that are included in the analyses do have follicles that are submarginal. We argue that the combined distinction in phylogenetic position, morphology, and host identity (being restricted to ostraciids) relative to Paradiscogaster species suggests the value in the recognition of Discogastroides for these two species.

Discogastroides ostracii (Yamaguti, 1934) Skrjabin & Koval, 1957

Synonyms: Discogaster ostracionis Yamaguti, 1934; Discogaster minor Yamaguti, 1934; Discogastroides hawaiensis Hanson, 1955; Discogastroides minor (Yamaguti, 1934) Skrjabin & Koval, 1957; Paradiscogaster ostracii (Yamaguti, 1934) Cribb, Anderson & Bray, 1999.

Type-host: Lactoria diaphana (Bloch & Schneider), Roundbelly cowfish (Tetraodontiformes: Ostraciidae) [as Ostracion diaphanum Bloch & Schneider].

Other hosts: Ostracion cubicum Linnaeus, Yellow boxfish; Ostracion immaculatum Temminck & Schlegel, Bluespotted boxfish; Ostracion meleagris Shaw, Spotted boxfish (Tetraodontiformes: Ostraciidae).

Type-locality: Kuki, Mie Prefecture, Japan.

Other localities: Tsushima Islands; Misaki, Osaka Prefecture; Moroiso Bay, Kanagawa Prefecture; Tanegashima, Kagoshima Prefecture; Fukaura, Aomori Prefecture; Kushimoto, Wakayama Prefecture, Japan; Hawaii; off Heron Island, Queensland, Australia.

New material

Known hosts: Ostracion cubicum; Ostracion meleagris.

New localities: Off Lizard Island (14°40′S, 145°27′E), northern Great Barrier Reef, Queensland, Australia; Ningaloo Reef (23°12′S, 113°45′E), Western Australia.

Known locality: Off Heron Island, southern Great Barrier Reef, Queensland, Australia (23°27′S, 151°55′E).

Prevalence: Heron Island: O. cubicum, 25/60 (42%); O. meleagris, 7/14 (50%). Lizard Island: O. cubicum, 3/18 (17%). Ningaloo Reef: O. cubicum, 6/6 (100%); O. meleagris, 3/3 (100%).

Deposition of specimens: 92 voucher specimens (G242543–634), including hologenophores, lodged in the QM; 58 voucher specimens (V13729–86), including hologenophores, lodged in the WAM.

Molecular sequence data: cox1 mtDNA, PZ518171–82; ITS2 rDNA, PZ510922–25; 28S rDNA, PZ513253–56.

Remarks

Specimens of Discogastroides ostracii were collected from O. cubicum and O. meleagris from off Heron Island and Lizard Island on the Great Barrier Reef and from Ningaloo Reef. The collections from Ningaloo Reef and Lizard Island represent new localities, with the Ningaloo Reef specimens significantly expanding the range of this species into the Indian Ocean.

Molecular data were generated for hologenophore specimens from all host/locality combinations. The cox1 data for these specimens form three clades (Heron Island only, Lizard Island + Ningaloo Reef, and Ningaloo Reef only) that differ at 8–15 bp (p-distance = 1.69–3.17%); minor intraclade variation was found at 0–4 bp (p-distance = 0–0.84%). Notably, two of the cox1 clades were found at Ningaloo Reef, but such a situation (multiple cox1 lineages found in a single location) is not unique for trematodes in the region. Multiple lineages of Preptetos laguncula Bray & Cribb, 1996, Transversotrema iapeti Cutmore, Corner & Cribb, 2023, and Emprostiotrema fusum (Goto & Ozaki, 1929) Cianferoni & Ceccolini, 2021 have been reported at Lizard Island (Bray et al. Reference Bray, Cutmore and Cribb2022; Cutmore et al. Reference Cutmore, Corner and Cribb2023; Huston et al. Reference Huston, Cutmore, Cribb, Sasal and Yong2025). ITS2 sequence data representing the three cox1 clades were identical except for some ambiguous base positions, and 28S data for each of the three cox1 clades differed at 2 bp (p-distance = 0.15%). Given the identical hosts and indistinguishable morphology of the hologenophore samples, we consider all of the base differences described above as being within the range of intraspecific variation.

The new specimens collected during this study almost completely agree with the description provided by Cribb et al (Reference Cribb, Anderson and Bray1999), who reported this species from O. cubicum and O. meleagris from off Heron Island. Cribb et al (Reference Cribb, Anderson and Bray1999) noted that they could not find any significant difference between the material from the Great Barrier Reef and specimens reported from Japan by Yamaguti (Reference Yamaguti1934) and Machida (Reference Machida1971); however, we draw attention to one subtle difference. Specimens from Australia have a distinctly medial genital pore, whereas the figures of both Yamaguti (Reference Yamaguti1934) and Machida (Reference Machida1971) suggest a slightly dextral genital pore. Notably, however, although Yamaguti (Reference Yamaguti1934) did not mention the position of the genital pore for Discogastroides ostracii, he did describe it as medial in his description of Discogaster (with Discogaster ostracionis proposed as the type species). We think that the subtle difference in the illustrated position of the genital pore may be attributable to specimen flattening. Although Yamaguti (Reference Yamaguti1934) does not specifically mention that these specimens were flattened, it was a common mounting technique during that period and, as understood now, may significantly alter overall body shape and the positions of organs.

Yamaguti (Reference Yamaguti1934) described a second species of Discogaster, D. minor, based on specimens from the same individual L. diaphana from which the specimens of D. ostracii were collected. He noted that the two species had essentially identical morphology, except for the size of the eggs, 48–57 × 24–27 for D. ostracii vs 33–36 × 18–21 for D. minor, and the extent of the caecal termination; Yamaguti considered the two species so similar that he did not provide a description or an illustration for Discogaster minor. Machida (Reference Machida1971) reported specimens from O. cubicum (as O. tuberculatus) and re-examined Yamaguti’s type-material, noting that while there was a difference in egg size, it was likely related to level of maturity. Shimazu and Kamegai (Reference Shimazu and Kamegai1990) also re-examined Yamaguti’s type-material, as well as reported and unreported specimens held in the Meguro Parasitological Museum (all from Japanese waters), and concluded that the differences in egg size fell within the range of intraspecific variation, and synonymised Discogaster minor. Cribb et al (Reference Cribb, Anderson and Bray1999) agreed with Shimazu and Kamegai (Reference Shimazu and Kamegai1990) and subsequently also synonymised Discogastroides hawaiensis Hanson, Reference Hanson1955 with D. ostracii. We measured multiple eggs from each of 54 specimens from Heron Island, Lizard Island and Ningaloo Reef (all paragenophores of specimens sequenced); eggs for these new specimens measure 29–44 × 15–25 (34 × 19). While the upper range of egg size is greater than that reported for Discogaster minor by Yamaguti, we found no specimens with eggs within the range reported in the original description of D. ostracii (48–57 × 24–27).

Regarding the gross morphology of D. ostracii, we also note some differences in the overall size of the specimens between the various reports of this species. In the original description, Yamaguti (Reference Yamaguti1934) provided measurements of the holotype only; the specimen, from Lactoria diaphana, was 1,120 μm long. For Discogaster minor, Yamaguti (Reference Yamaguti1934) reported a body length of 990–1,070 μm. Hanson (Reference Hanson1955) described specimens from Ostracion meleagris from Hawaii as Discogastroides hawaiensis, reporting a body length of 1,197–1,679 μm. Machida (Reference Machida1971) reported specimens of Discogastroides ostracii from O. cubicum from Japan, with a much larger body length of 1,590–3,060 μm. Shimazu and Nagasawa (Reference Shimazu and Nagasawa1985) and Machida and Kuramochi (Reference Machida and Kuramochi1999) both reported this species from Japanese waters (from O. cubicum and O. immaculatum) but did not provide any measurements of their specimens. Cribb et al (Reference Cribb, Anderson and Bray1999) reported specimens from O. cubicum and O. meleagris from Australia, reporting a body length of 822–1,566 μm. We measured 42 specimens from Heron Island, Lizard Island and Ningaloo Reef (all paragenophores of specimens sequenced), and found the specimens to be 658–1,292 μm long. Thus, our specimens generally agree with those of Yamaguti (Reference Yamaguti1934) and Cribb et al. (Reference Cribb, Anderson and Bray1999), are slightly smaller than those of Hanson (Reference Hanson1955), and are much smaller than those of Machida (Reference Machida1971).

Ultimately, determining whether specimens from L. diaphana, O. cubicum and O. immaculatum from Japan, specimens from O. meleagris from Hawaii, and specimens from O. cubicum and O. meleagris from Australia are conspecific will rely on generating novel molecular data for specimens from Japan, Hawaii and elsewhere in the tropical Indo-West Pacific in between these locations. Notably, as part of this study, we examined one O. cubicum, two O. immaculatum and seven O. meleagris in Japan (from off Okinawa and Minabe) and found no specimens of D. ostracii.

Discogastroides macrostomus Shimazu & Kamegai, 1990

Synonym: Paradiscogaster macrostomus (Shimazu & Kamegai, 1990) Cribb, Anderson & Bray, 1999.

Type-host: Ostracion immaculatum Temminck & Schlegel, Bluespotted boxfish (Tetraodontiformes: Ostraciidae).

Type-locality: Sagami Bay, Kanagawa Prefecture, Japan.

Other hosts: Ostracion cubicum Linnaeus, Yellow boxfish; Ostracion meleagris Shaw, Spotted boxfish (Tetraodontiformes: Ostraciidae).

Other locality: Off Heron Island, Queensland, Australia

New material

Known hosts: Ostracion cubicum; Ostracion meleagris.

New localities: Ningaloo Reef (23°12′S 113°45′E), Western Australia; off Mizugama (26°21′N, 127°44′E); Zanpa Lagoon (26°26′N, 127°42′E), Okinawa Prefecture, Japan.

Known locality: Off Heron Island, southern Great Barrier Reef, Queensland, Australia (23°27′S, 151°55′E).

Prevalence: Heron Island: O. cubicum, 2/60 (3%); O. meleagris, 1/14 (7%). Ningaloo Reef: O. cubicum, 6/6 (100%); O. meleagris, 1/3 (33%). Mizugama: O. meleagris, 1/4 (25%). Zanpa Lagoon: O. cubicum, 1/1 (100%).

Deposition of specimens: Seven vouchers (G242635–41), including hologenophores lodged in the QM; two vouchers (MPM Coll. No. 25530–31), including one hologenophore lodged in the MPM; seven vouchers (V13787–93), including hologenophores lodged in the WAM.

Molecular sequence data: cox1 mtDNA, PZ518183–89; ITS2 rDNA, PZ510926–27; 28S rDNA, PZ513257–58.

Remarks

Specimens of Discogastroides macrostomus were collected from O. cubicum and O. meleagris from off Heron Island, Ningaloo Reef, and Okinawa, Japan. The collections from Ningaloo Reef and Okinawa represent new localities for this species. As part of this study, we examined two Ostracion immaculatum from Japanese waters but found no zoogonids.

Molecular data were generated for hologenophore specimens from Ningaloo Reef and Okinawa. The cox1 data for these specimens form a single clade, with intraclade variation found at 1–23 bp (p-distance = 0.21–4.85%); notably, the Japanese samples are paraphyletic relative to the Ningaloo specimens. Samples from Ningaloo Reef and Japan differed at just 1 bp (p-distance = 0.25%) in the ITS2 dataset and 4 bp (p-distance = 0.31%) in the 28S dataset. We consider the differences to be within the range of intraspecific variation, especially over such large geographic distances.

The new specimens agree morphologically with the emended description of Cribb et al (Reference Cribb, Anderson and Bray1999), possessing the characteristic large tegumental spines and strikingly submedial genital pore. Similar to the samples reported by Cribb et al (Reference Cribb, Anderson and Bray1999), the new samples were slightly smaller than those reported by Shimazu and Kamegai (Reference Shimazu and Kamegai1990), but still had overlapping ranges with the type series (782–927 long vs 790–1780 long). Notably, the samples reported by Cribb et al (Reference Cribb, Anderson and Bray1999) had smaller eggs than those reported by Shimazu and Kamegai (Reference Shimazu and Kamegai1990) (25–30 vs 32–40 long). Our new specimens agreed with the smaller egg size of Cribb et al (Reference Cribb, Anderson and Bray1999) (25–32 long), but given these measurements were derived from a series that included specimens from Japan, we consider that difference in egg size as within an intraspecific range.

Parayamagutia Machida, 1971

Diagnosis

See Machida (Reference Machida1971).

Type and only species: Parayamagutia ostracionis Machida, 1971.

Remarks

Cribb et al. (Reference Cribb, Anderson and Bray1999) noted some striking similarities between Discogastroides macrostomus and Parayamagutia ostracionis, especially in the body spination, extent of the uterus, and the position of the genital pore. Given that Parayamagutia ostracionis is the sister taxon to species of Discogastroides in all analyses, the recognition of Parayamagutia as valid or as a synonym of Discogastroides is partially subjective. Based on the available data, we think that the differences between Parayamagutia and Discogastroides are more significant than the similarities. First, the body shape is strikingly different; species of Discogastroides are pyriform, and Parayamagutia ostracionis is oval. Second, the position of the vitellarium is distinct between the two genera: for Discogastroides, it is at the level of the ventral sucker and immediately anterior to it, and for Parayamagutia, it is at the level of the oral sucker and immediately posterior to it. Third, the uterus is significantly more extensive in Parayamagutia, reaching anteriorly past the posterior margin of the oral sucker, whereas in species of Discogastroides the uterine coils only reach anteriorly to the level of the genital pore. Ultimately, the discovery of more species in this clade will inform whether such distinctions represent genus or species level differences, but until such time we think it most appropriate to consider the two genera distinct.

Parayamagutia ostracionis Machida, 1971

Type-host: Ostracion cubicum Linnaeus, Yellow boxfish (Tetraodontiformes: Ostraciidae) [as Ostracion tuberculatus Linnaeus].

Type-locality: Off Tsushima Island, Japan.

Other locality: Off Heron Island, Queensland, Australia

New material

New host: Ostracion meleagris Shaw, Spotted boxfish (Tetraodontiformes: Ostraciidae).

Known host: Ostracion cubicum.

New locality: Off Lizard Island, northern Great Barrier Reef, Queensland, Australia (14°40′S, 145°27′E).

Known locality: Off Heron Island, southern Great Barrier Reef, Queensland, Australia (23°27′S, 151°55′E).

Prevalence: Heron Island: O. cubicum, 1/60 (2%); O. meleagris, 1/14 (8%). Lizard Island: O. cubicum, 1/18 (11%).

Deposition of specimens: One voucher specimen (G242642) lodged in the QM.

Molecular sequence data: cox1 mtDNA, PZ518215–16; ITS2 rDNA, PZ510928–29; 28S rDNA, PZ513252.

Remarks

Just three specimens of Parayamagutia ostracionis were collected during this study, two from O. cubicum and one from a new host, O. meleagris. This is just the second report of Parayamagutia ostracionis since its description by Machida (Reference Machida1971), with Cribb et al (Reference Cribb, Anderson and Bray1999) reporting a single specimen collected from an examination of 18 O. cubicum from off Heron Island. Like Cribb et al (Reference Cribb, Anderson and Bray1999), we found this species to be rare, finding just a single specimen from 60 O. cubicum at Heron Island, a single specimen from 14 O. meleagris at Heron Island, and a single specimen from 18 O. cubicum from off Lizard Island. No infections were found from the examination of six O. cubicum and three O. meleagris from Ningaloo Reef, one O. cubicum and seven O. meleagris from off Okinawa, or eight O. meleagris from French Polynesia.

The two specimens collected from O. cubicum were used in their entirety for genetic characterisation, and no morphological data are available for these specimens. The single specimen from O. meleagris was processed as a hologenophore, but the molecular portion failed to amplify; thus, there is a partial mount but no genetic data for this specimen. The two sequenced specimens of Parayamagutia ostracionis, one each from Heron Island and Lizard Island, differed at 10 bp (p-distance = 2.11%) in the cox1 dataset, and had identical ITS2 sequence data; just a single 28S sequence was generated for this species, from Lizard Island.

The limited morphological data obtainable from the single partial specimen matched previous descriptions by Machida (Reference Machida1971) and Cribb et al (Reference Cribb, Anderson and Bray1999). The new specimen possesses a highly extensive uterus, distinctive to this species; in the new specimen the uterus reaches anteriorly to the midpoint of the oral sucker. The eggs in the new specimen measured 28–29 × 15–18, equivalent to egg measurements reported by Machida (Reference Machida1971) (25–31 × 12–17) and only slightly smaller than those reported by Cribb et al (Reference Cribb, Anderson and Bray1999) (30–33 × 17–19). Finally, the oral sucker measured 208 × 226, similar to the reports by both Machida (Reference Machida1971) (182–284 × 228–294) and Cribb et al (Reference Cribb, Anderson and Bray1999) (218 × 212).

Scaticola n. g.

Diagnosis

Body short, variably fusiform. Tegument spinose; spines distinct to posterior end of body. Oral sucker terminal, transversely oval or infundibuliform. Ventral sucker small, not strongly muscular, usually without evident cavity but may have aperture consisting of transverse slit, in mid-body. Prepharynx short. Pharynx small, subglobular or transversely oval. Intestinal bifurcation in forebody. Caeca short, saccular, divergent, terminate in forebody or close to anterior margin of ventral sucker. Testes subglobular, opposite to oblique, at level of ventral sucker or overlapping into anterior hindbody. Cirrus-sac oval, recurves slightly distally before entering common genital atrium. Internal seminal vesicle bipartite. Pars prostatica long, straight, prominent. Eversible ejaculatory duct short, straight. Genital pore median, anterior to anterior margin of ventral sucker. Ovary subglobular, at level of testes. Canalicular seminal receptacle present. Laurer’s canal present, opens dorsally close to posterior end of body; no specialisation associated with opening. Vitelline follicles in separate lateral fields; fields at level of caeca or posterior portion of oesophagus. Uterus extensive, fills most space posterior to vitelline follicles. Eggs tanned, operculate. Excretory pore terminal. Excretory vesicle V-shaped; short stem present. In intestine of scatophagid fishes.

Type-species: Scaticola lesteri (Bray, 1982) n. comb.

Other species: Scaticola sprenti (Bray, 1982) n. comb.

ZooBank LSID: urn:lsid:zoobank.org:act:DE67086C-A611-4072-A40A-C9C7D775F0F0.

Etymology: This genus is named for the common name of the fishes it infects (Scats) and the Latin cola (dweller or inhabitant). The generic name is masculine.

Remarks

Scaticola is proposed to accommodate two known species currently assigned to Bacciger, B. lesteri and B. sprenti. Bray (Reference Bray1982) described both species from an unidentified mugilid from Moreton Bay, Australia. In their review of Australian faustulids, Cribb et al (Reference Cribb, Anderson and Bray1999) re-reported both species from the striped scat, Selenotoca multifasciata (Richardson) (Scatophagidae), from the type locality, noting that the host identification of a mugilid by Bray (Reference Bray1982) was likely erroneous based on further examination of both host groups in Moreton Bay. Additionally, Greiman et al (Reference Greiman, Tkach, Pulis, Fayton and Curran2014) also reported B. sprenti from S. multifasciata, rather than a mugilid, from Harvey Bay, approximately 200 km north of Moreton Bay. We agree with the speculation of Cribb et al (Reference Cribb, Anderson and Bray1999); since their 1999 study, we have found these two species in 15/15 S. multifasciata and 0/77 mugilids examined from Moreton Bay.

Olson et al (Reference Olson, Cribb, Tkach, Bray and Littlewood2003) incorporated sequence data for B. lesteri in their analyses of the Digenea, with the species resolving with Neoparantorchis pomacanthi and Trigonocryptus conus, within the Zoogonidae. Multiple phylogenetic studies since have demonstrated the same relationship for this taxon (e.g., Cutmore et al Reference Cutmore, Miller, Bray and Cribb2014; Duong et al Reference Duong, Martin, Cutmore and Cribb2025; Kremnev et al Reference Kremnev, Gonchar, Uryadova, Krapivin, Skobkina, Gubler and Krupenko2023; Pérez Ponce de León and Hernández-Mena Reference Pérez Ponce de León and Hernández-Mena2019; Sun et al Reference Sun, Bray, Yong, Cutmore and Cribb2014), and more recently a range of studies have shown it to be only distantly related to other species of Bacciger (Curran et al Reference Curran, Warren and Bullard2022; Cutmore et al Reference Cutmore, Bray and Cribb2018; Le et al Reference Le, Jeung, Cho and Choi2024). Curran et al (Reference Curran, Warren and Bullard2022) considered B. lesteri and B. sprenti as incertae sedis. Here, we assign the two species to a new genus, within the Antorchiinae.

Within the Antorchiinae, Scaticola is morphologically most similar to the concepts of Neoparantorchis and Parantorchis. Scaticola lesteri and S. sprenti differ from the concept of Neoparantorchis in the position of the oral sucker (terminal vs subterminal), position of the ventral sucker (middle third of body vs posterior third of body), the extent of the uterus (coils extending anteriorly past the ventral sucker vs restricted to the region posterior to the ventral sucker), having slightly fewer vitelline follicles per lateral cluster and the shape of the excretory vesicle (V-shaped vs Y-shaped). Scaticola lesteri and S. sprenti differ from the concept of Parantorchis in the position of the oral sucker (terminal vs subterminal), caeca length (terminating anterior to testes vs reaching posteriorly beyond middle of testes), the extent of the uterus (coils extending anteriorly past the ventral sucker vs restricted to the region posterior to the ventral sucker) and the shape of the excretory vesicle (V-shaped vs Y-shaped). In the new 28S analyses, Scaticola lesteri and S. sprenti form a strongly supported clade sister to a clade of Discogastroides, Paradiscogaster, and Parayamagutia; this large clade is sister to the clade of Neoparantorchis and Parantorchis.