Schistorchis carneus Lühe, 1906

Species Schistorchis carneus Lühe, 1906 View in CoL

(Syns.: Pleorchis oligorchis Johnston, 1913 ; Schistorchis oligorchis View in CoL [ Johnston, 1913] Yamaguti, 1942)

Host(s): White-spotted puffer, Arothron hispidus (Linnaeus) ( Tetraodontidae ).

Locality: Northern Red Sea, off Safaga, Southern Hurghada, Egypt.

Collection dates: 6/Apr/2020 & 12/Jan/2022.

Site of infection: Mid-intestine.

Deposited material:Parasitology Laboratory,Zoology Department, Faculty of Science, South Valley University; Collector YFMK & AM, Vouchers (n = 14) (SVU) 2023.20.Sc01–Sc14.

Prevalence: 33.3% (2 out of 6 host infected).

Worm burden: 5 &13 worms.

Mean intensity: 9.0 (18/2).

Relative density/abundance: 3.5 (18/6).

Representative DNA sequences: ITS2 rDNA (accession no. PP320859) and 28S rDNA (accession no. PP320857).

Supplemental description

Figs. (1–3)

[Based on 14 mature specimens]. Worms blood red. Body plump, elongated to sub-elliptical with almost parallel margins; maximum width near equatorial region. Tegument thick, aspinose, with numerous transverse folds along entire body; dis-concentric arrangement and several narrow furrows. Tegumental folds in early mature specimens domed-like, large structures. Anterior end rounded; posterior end more broadly rounded, sometimes truncated. Forebody shorter than hind body, gradually narrows near anterior end; hindbody wider, gradually narrows in posterior 1/5 of body. Pre-oral lobe very short. Oral sucker spherical, glandular, ventro-subterminal, of complex type; with margins of anterior musculature notched or indented; situated on top of a rather globular cephalic end that is distinctive by a slight constriction just posterior to oral sucker. Mouth inconspicuous, rounded, directed anteroventrally; surrounded by strong semi-circular muscle fibers, sometimes covers mouth opening. Sphincter partially U-shaped, muscular, half-encircles oral aperture postero-ventrally. Ventral sucker pre-equatorial, at junction of first and second quarters of body, spherical, sessile, unspecialized, less than half size of oral sucker; situated on top of a large body protrusion. Suckers separated by a short distance to somewhat contiguous. Pre-pharynx absent. Pharynx transversally elliptical, often longitudinally compressed, smaller than suckers, weakly developed, not papillate, often overlapped by oral sucker. Esophagus not evident. Intestinal bifurcation posterior to mid-forebody, midway between suckers, separated from ventral sucker by a small distance. Ceca two, simple, wider anteriorly, slightly arcuate posteriorly, equal in length, extend medially to lateral margins; terminate at posterior extremity in small, separate ani. Each anus lateral, slit-like, narrow, subterminal.

Testes 11, rarely fewer (12 specimens with 11 testes distributed 5 left and 6 right; 1 specimen with 10 testes distributed 4 left, and 6 right; 1 specimen with 9 testes distributed 4 left and 5 right). Testes arranged in two straight or zigzag rows, smooth to slightly indented, sub-elliptical to sub-oval to irregular, equal or differing in size, contiguous or separated by a short distance. Testes occupy area posterior to ventral sucker, reach to anterior region of second half of body. Pre-testicular distance anterior one-third of body length; post-testicular space posterior twofifths of body. Seminal vesicle saccate, conspicuous, large, transversely sub-spherical to elliptical, unipartite, with narrower distal portion; confined to area between ventral sucker and beginning of second third of body, majority dextral to midline. Pars prostatica tubular, long, thin-walled, parallel to ventral sucker to mid-sinistral rim of latter. Prostatic cells well-developed, free in parenchyma, surrounds entire pars prostatica from junction with seminal vesicle to beginning of hermaphroditic duct. Genital pore median, opening immediately anterior to ventral sucker.

Ovary round to oval, smooth, dextro-submedian, at end of first third of body; either entirely pre-testicular or at level of anteriormost testis, confined between seminal vesicle and testes. Seminal receptacle conspicuous, canalicular, large, broadly saccate, variable in size, sinistral or postero-sinistral to ovary. Oötype median, sinistral to ovary; surrounded by a large mass of Mehlis’ gland cells. Laurer’s canal indistinct. Vitellarium follicular, extends along lateral margins from mid-level of pharynx or mid-level of ventral sucker to terminating near posterior extremity. Vitelline follicles numerous, small, sub-spherical to oval, dorsal, ventral, lateral to ceca. Right and left transverse vitelline ducts unite immediately anterior to ovary and form a small median vitelline reservoir that opens into oötype. Uterus pre-testicular, short, coiled, inter-cecal, located in anterior hindbody; restricted to area among ovary, seminal vesicle, oötype complex, anteriormost left testis, and genital pore. Post-uterine distance very long; about two-thirds of body length. Metraterm tubular, short, wide; extends parallel to pars prostatica, then unites with its distal portion to form a small cylindrical hermaphroditic duct that extends anterodorsally to ventral sucker. Hermaphroditic duct less than half length of Pars-prostatica. Eggs numerous, operculated, oval, thin-shelled, nonfilamented.

Excretory vesicle I-shaped, narrower posteriorly, median, extends to the second third of body, terminates at level of anteriormost testes. Two excretory ducts cross ceca ventrally and runn somewhat sinuously as far as oral sucker. Excretory pore terminal, elongated, narrow, slit-like.

Variability in adult specimens. Applying the coefficient of variation on the most significant morphometric features of specimens of S. carneus revealed a wide range of variability ranging among less variable (CV<10%), moderate (10%<CV<15%), and more variable (CV>20%) features as shown in Table (3). A few metric characters showed low CV with a significant correlation with body length (r>0.59, p<0.05) particularly the pre-bifurcal distance, egg size and oral sucker length as percentages of body length. Other metric characters exhibited low CV with non-significant correlation to the body length (r <0.39, p>0.05), including ratios of mean testis length, pre-ovarian distance, post-uterine distance, and post-testicular distance as percentages of body length in addition to sucker width ratio. Such characteristics are functional in taxa differentiation and comparisons. Out of 26 morphometric characters with mild CV, 14 showed a significant correlation with body length (r>0.50, p<0.05), including body width, forebody length, oral sucker length, pharynx width, ventral sucker length, mean testis length, mean testis width, pre-genital pore distance, pre-vitelline distance, pre-ovarian distance, pre-testicular distance, post-testicular distance, post-uterine distance, and the ratio of ventral sucker length to body length. Concerning morphometric characters with high CV, the ratios of pre-oral distance, distance between suckers, and distance from the intestinal bifurcation to the ventral sucker as percentages of body length were notably greater, respectively. Furthermore, ratios of the intestinal bifurcation to ventral sucker distance, hermaphroditic duct length, ventral sucker to ovary distance, and pre-oral distance as percentages of body length showed a significant correlation with body length (r>0.50, p<0.05).

Remarks. Our specimens were identified as belonging to the Megaperidae within the Apocreadioidea based on following key diagnostic combination of features; parasitic in the alimentary canal of marine fish, possessing mouth at anterior extremity, presence of both suckers and pharynx, the union of male and female terminal ducts forming a hermaphroditic duct without a hermaphroditic sac, an unarmed tegument, the absence of a prepharynx, an I-shaped excretory vesicle, and testes positioned in the hindbody ( Bray 2005 a, 2005b; Cribb 2005; Jones 2005; Blend et al. 2017). The positioning of the genital pore anterior to the ventral sucker and the presence of an oral sucker with a partial U-shaped sphincter half encircling the oral aperture postero-ventrally align our specimens with the Schistorchiinae ( Bray 2005b; Cribb 2005; Pulis et al. 2014; Blend et al. 2017). Our material keyed out to Schistorchis , based on having more than five testes and a highly glandular oral sucker ( Cribb 2005; Blend et al. 2017; Magro et al. 2023). Based on specific characteristics, including a relative egg size less than 90 × 55 µm, the anterior extent of the uterus reaching to the level of or near the anterior margin of the ventral sucker, a single anterolateral circumoral lobe with varying degrees of anterior indentation, an ovary about equal to or slightly larger than the testes, a seminal receptacle located posterior or lateral to the ovary, testes arranged in two rows, vitelline follicles noticeably smaller than the testes and a seminal vesicle encroaching and/or intervening into the space between the ovary and ventral sucker ( Blend et al. 2017) as well as possessing similar measurements, morphometric percentages, and ratios of observed features ( Table 3 View TABLE 3 ), we have identified our material as S. carneus .

A review of previous records of S. carneus alongside our material revealed some variations in morphological features, the most notable being the anterior extent of the vitellarium, the appearance of the male genital system, the shape of ceca, the configuration of the pre-oral lobe, and the arrangement of the testes. (i) The anteriormost extent of the vitellarium varied from being at the mid-level of the pharynx ( Blend et al. 2017; Present study), beginning at the level of the ovary or seminal vesicle ( Lühe 1906; Madhavi et al. 1986), or at the level of the ventral sucker ( Johnston 1913; Hafeezullah 1981; Blend et al. 2017, fig. 16; Present study). (ii) The seminal vesicle was either located in the median line just posterior to the ventral sucker ( Lühe 1906, figs. 11&12) or appeared as an elongated, pear-shaped small sac extending posterior to the ventral sucker from the left side and narrowing anteriorly to form a short pars prostatica ( Johnston 1913, fig. 11; Madhavi et al. 1986, fig. 6). The present study, however, describes the seminal vesicle as a simple, saccate, wide, oblong to clavate pouch, transversally sub-spherical to elliptical, and extending dextral to the midline. Such a result is consistent with Blend et al. (2017), and illustrations of Hafeezullah (1981, fig. 6) and Johnston (1913, fig. 38). (iii) The ovary and majority of the proximal seminal vesicle were situated dextral and opposite to the sinistro-medial uterus ( Blend et al. 2017; Present study), whereas other investigations revealed that the ovary was either medial ( Hafeezullah 1981) or dextro-submedian ( Lühe 1906; Johnston 1913; Madhavi et al. 1986). (iv) In the present study, the pars prostatica was longer than the hermaphroditic duct in agreement with Blend et al. (2017). In contrast, the hermaphroditic duct was equal to or longer than the pars prostatica in other described forms ( Lühe 1906, figs. 11 & 12; Johnston 1913; Hafeezullah 1981, fig. 6; Madhavi et al. 1986, fig. 6). (v) The intestine generally lacked anterior ceca ( Lühe 1906; Hafeezullah 1981; Madhavi et al. 1986; Blend et al. 2017; Present study) though Johnston (1913) occasionally observed a somewhat H-shaped structure there. (vi) Only specimens of Madhavi et al. (1986) possessed furrows in the pre-oral lobe region. (vii) Magro et al. (2023) clarified that the testes of S. carneus could be arranged in either a single median row or two submedian rows, whereas in previous studies as well as ours, we consistently observed the arrangement in two submedian rows. Blend et al. (2017) suggested that their material might represent an undescribed species based on the further anterior extent of the vitellarium up to the level of the mid-pharynx, the longer pars prostatica relative to the hermaphroditic duct, and the considerably distant locality of their specimens compared to other records of S. carneus (Northern Red Sea, off Sharm El-Naga, Makadi Bay, Southern Hurghada and off Safaga, Egypt vs list the localities of other records of S. carneus ), in addition to a combination of minor morphological differences (a more conspicuous and larger seminal vesicle extending further dextral to the mid-line, a slightly longer egg, and a dextral ovary with the majority of the proximal seminal vesicle vs a rudimentarily described seminal vesicle of almost wholly median position, a marginally shorter egg, and a medial ovary or a little to the right of the median line. However, Blend et al. (2017) opted to identify their specimens as S. carneus , waiting for further evidence. Observed differences were considered intraspecific variations likely due to locality-induced variability, degree of worm maturity, and postfixation treatments ( Blend et al. 2017).

Regarding variations in morphometric percentages/ratios, the present specimens and materials from Blend et al. (2017) and Lühe (1906) showed a wide range of variability, which in turn, overlaps and causes notable confusion when compared to other described forms of S. carneus (see Table 3 View TABLE 3 ). However, some of these forms display distinct and relatively consistent morphometric values: Specimens of Lühe (1906) exhibited the highest values in pharynx/ ventral sucker width ratio. Material from Hafeezullah (1981) specimens had the largest body length, distance between suckers, oral sucker length to body length ratio, and sucker length ratios. Specimens of Madhavi et al. (1986) possessed the largest body length, the highest values in forebody length as a percentage of body length, and the largest mean testis/ovary length ratio. Conversely, these specimens had the lowest ratios for oral sucker length, pharynx length, mean testis length, post-testicular distance, and post-vitelline distance relative to body length. Specimens of Magro et al. (2023) from A. stellatus off Lizard Island had the highest values in ventral sucker length, mean testis length, distance from the ventral sucker to the anterior testis, pre-testicular and pre-vitelline distance, post-vitelline distance as percentages of body length, and the pharynx/ventral sucker width ratio. Moreover, they exhibited the smallest values in the body length, and mean testis/ovary length and width ratios. Specimens of Magro et al. (2023) from the same host A. stellatus off Moreton Bay had the highest value in mean testis/ovary width ratio and the lowest distance between suckers. Concerning specimens of Magro et al. (2023) from A. manilensis off Mackay Harbour, they showed the highest values in ratio of pre-testicular distance to body length and the highest sucker width ratio.

According to host-parasite data, the described forms of S. carneus are distributed among three distinct and geographically differentiated regions: the Australian group ( Johnston 1913; Magro et al. 2023), the Sri Lankan/ Indian group ( Lühe, 1906; Hafeezullah 1981; Madhavi et al.1986), and the Egyptian group ( Blend et al. 2017; Present study). Concerning hosts, all forms of S. carneus are confined among five tetraodontid hosts: lunartail puffer, L. lunaris ( Madhavi et al.1986) ; stellate puffer, A. stellatus ( Lühe, 1906; Magro et al. 2023); white-spotted puffer, A. hispidus ( Johnston 1913; Hafeezullah 1981; Blend et al. 2017; Present study); narrow-lined puffer, A. manilensis ( Magro et al 2023); and an unidentified black toad-fish ( Johnston 1913). The first three hosts are widely distributed near the shores of Australia, India, the Red Sea, and neighboring regions, while A. manilensis is restricted to off Australia and neighbouring zones ( Philippines, Samoa, Ryukyu Islands, New South Wales, and Tonga) ( Froese & Pauly 2024).

Tegumental folding, ridges, and furrows in S. carneus provide significant stretching ability, which in turn, enhances the worm's surface area for absorbing micro-molecular nutrients (see Parkening & Johnson 1969; Nollen et al. 1973; Panyarachun et al. 2010; Swarnakar et al. 2014). The presence of a few folds encircling the oral sucker, near the ventral sucker, and the genital pore can be attributed to the location within the forebody where the worm attaches to the host intestine ( Swarnakar et al. 2014). At such attachment sites, suckers come into direct contact with the plug of host tissue formed by the fluke ( Hoole & Mitchell 1981; Ashour 1995). Furthermore, the absence of sensory cilia, papillae, and spines in the tegument overlying the two suckers is probably an adaptation that allows the suckers to make a smooth seal with the substrate prior to suction ( Bennett 1975; Parkening & Otubanjo 1985).

Molecular analysis

Genetic distances

( Tables S2–S View TABLE 2 4)

Two matrices with the ITS2 rDNA and the partial 28S rDNA in addition to a supplementary matrix of cox 1 mtDNA were constructed. The 28S rDNA region was aligned with 55 retrieved sequences representing all the available and appropriate taxa of megaperids: 34 sequences from four species representing the Apocreadiinae Skrjabin, 1942 , four sequences from three species representing the Megaperinae Manter, 1934 and 14 sequences from the Schistorchiinae with four sequences representing three species from the Haploporidae Nicoll, 1914 , for outgroup comparison. The final alignment of the 28S rDNA datasets included 57 sequences, resulting in a 1,264 bp alignment, comprising 781 bp (61.9%) conserved sites and 482 bp (38.1%) variables, with 387 bp (30.6%) prism-informative sites, 95 bp (7.5%) singleton sites, and a genetic divergence range of 0%–13.2% within megaperids. The ITS2 rDNA comparisons comprised 21 taxa of schistorchiines: eight sequences representing three species from Paraschistorchis Blend, Karar & Dronen, 2017 ; five sequences representing two species of Blendiella Magro, Cutmore, Carrasson & Cribb, 2023 ; four sequences representing two species from Schistorchis ; three sequences from one taxon within Lobatotrema Manter, 1963 ; and one sequence of Sphincteristomum Oshmarin, Mamaev & Parukhin, 1961 . The final alignment ITS2 rDNA datasets comprised 22 sequences, yielding a total of 425 aligned positions, with 304 bp (81.6%) conserved sites and 78 bp (18.4%) variables. Of these, 73 bp (17.2%) were parsimony informative sites, five bp (1.2%) were singleton sites, and genetic divergence ranged from 0.0%–11.3%.

The Egyptian and Australian forms of S. carneus were identical, with no nucleotide differences in the 1,264 bp long 28S rDNA sequences. Differences between S. carneus and Schistorchis skrjabini Parukhin, 1963 were 2.7% (34 bp). Within the 425 bp ITS2 rDNA alignment, the Egyptian form of S. carneus ex. A. hispidus differed by 0.002% (1 bp) and 0.005% (2 bp), respectively, at different nucleotide sites from the two Australian forms ( S. carneus ex. A. stellatus and S. carneus ex. A. manilensis ). The Australian forms of S. carneus ex. A. stellatus differed by 0.007% (3 bp) from those gathered from A. manilensis . Additionally, S. skrjabini exhibited a variability of 2.6% (11 bp) from the Egyptian form of S. carneus ex. A. hispidus and the Australian form, S. carneus ex. A. stellatus , while it differed by 2.8% (12 bp) from the Australian form, S. carneus ex. A. manilensis .

Phylogenetic analysis

( Fig. 4 View FIGURE 4 )

The phylogenetic analysis of the 28S rDNA dataset resulted in the ingroup taxa of the Megaperidae forming a monophyletic clade to the exclusion of the outgroup taxa ( Fig. 4 View FIGURE 4 ). Each of the three recognized subfamilies was resolved as highly supported monophyletic clades; the Schistorchiinae (BI=1; ML=95), the Megaperinae (1/100), and the Apocreadiinae (1/99). Schistorchiines and megaperines were resolved as sister clades in a strongly supported Schistorchiinae / Megaperinae clade (1/100), which in turn was resolved a sister to the Apocreadiinae . Taxa within the Schistorchiinae differentiated into two clusters: a weakly supported Lobatrema / Paraschistorchis clade (0.83/66) and a strongly supported clade (1/92) constituting Schistorchis , Sphincteristomum , and Blendiella . Taxa of Schistorchis clustered into a highly supported clade (1/100), which was resolved as a sister to the highly supported Sphincteristomum / Blendiella clade (1/100); specimens of S. carneus clustered in a 3-way polytomy (1/100). Although the Lobatrema / Paraschistorchis clade exhibited a weak support, it divided into two strongly supported clades, Lobatrema clade (1/100) and a Paraschistorchis one (1/100).

Remarks. The addition of several sequences from schistorchiine members ( Magro et al. 2023; Present study) showed the close association of the Schistorchiinae and the Megaperinae , indicating that megaperines are more closely allied with taxa of the Apocreadioidea than with the Lepocreadioidea Odhner, 1905 . These results follow the phylogenetic results of Pulis et al. (2014) that included the Megaperidae within the Apocreadioidea as a synonym of the Apocreadiidae and reduced its rank to the subfamily level. Blend et al. (2017) supported this conclusion, arguing that according to the "ICZN principle of priority," the correct family name should be Megaperidae , not Apocreadiidae . We are consistent in the reorganization of megaperines as a distinct subfamily within the Megaperidae , as proposed by Blend et al. (2017), following the consideration of Apocreadiidae as a junior synonym of Megaperidae ( ICZN, 2022; Blend et al. 2024; WORMS, 2024a). The closer relationship between the Megaperinae and the Schistorchiinae , compared to the Apocreadiinae , may be due to shared characteristics, in particular, possession of an anus/ani and their common association with marine tetraodontiform hosts. In contrast, apocreadiines have blind-ending ceca or a cyclocoel and parasitize the intestine of non-tetraodontiform marine and freshwater fishes. Additionally, megaperines can be differentiated from schistorchiines by possessing a pharynx with a lobed anterior rim and lacking a U-shaped sphincter half encircling the oral aperture ( Bray 2005b; Cribb 2005; Pulis et al. 2014; Blend et al. 2017).

Blend et al. (2017) reclassified Schistorchis (sensu lato) into three genera: Paraschistorchis , Plesioschistorchis Blend, Karar & Dronen, 2017 , and Schistorchis (sensu stricto) based on a combination of some well-considered generic differential features including the morphology of the oral sucker (either glandular or muscular), the nature of the cecal ends (either opening via separate ani or ending blindly), and the host type (either tetraodontiform and/ or perciform marine teleosts). The present phylogenetic analyses show a notable divergence between Schistorchis (sensu stricto) and Paraschistorchis taxa. In contrast, an obvious convergence with high support is observed among Schistorchis (sensu stricto), Sphincteristomum and Blendiella which reflect the validity of division of the expanded concept of Schistorchis by Blend et al. (2017). The most distinguishing synapomorphy among the latter three genera is the "complex" nature of the oral sucker; that is characterized by its larger size relative to body size, the presence of circumoral lobes, and a predominantly glandular appearance. Paraschistorchis was established to accommodate taxa having an oral sucker of a “simple” nature which is characterized by being smaller in size compared to that of a "complex" nature, lacking circumoral lobes, and being entirely muscular. By extension, it is also expected that members of Plesioschistorchis will phylogenetically group with Paraschistorchis and other schistorchiines having an oral sucker of "simple" nature.

Records of sequences of S. carneus showed some differences and overlap. i) Despite using a sufficiently long 28S rDNA region (1,264 bp), the Egyptian and Australian forms of S. carneus were resolved as a polytomy without any nucleotide differences in the 28S rDNA region. Such polytomous nodes reflect the limitations of the sequence data of records of S. carneus and the inability to display dichotomic clades with low support. In addition, these polytomies show the data genuinely support multiple, equally probable relationships, possibly due to rapid diversification, a lack of informative characters, or homoplasy (see Lewis et al. 2005). ii) A tiny difference in the ITS2 rDNA region resulted in the distribution of specimens of S. carneus into three highly supported and closely associated groups, apparently according to the type of the host (see Table S3 View TABLE 3 ). iii) The previously mentioned variations in morphological features and morphometric measurements among all forms further complicate the picture. All these observations may reflect that all sequences represent the same species (i.e., S. carneus ), which exhibits a wide range of intraspecific variations. Alternatively, they could represent extremely closely related taxa that lack resolution among their lineages within the ribosomal 28S rDNA and ITS2 rDNA regions (see Hershkovitz & Lewis 1996; Kuzmin et al. 2017), thus, suggesting potential cryptic species.

Species delimitation analysis

( Figs. 5 View FIGURE 5 , S 1 View FIGURES 1 & 2 )

The phylogenetic tree of the unrooted ITS2 rDNA resolved schistorchiines into two strongly supported clades (NJ=99); the Paraschistorchis clade and a larger cluster comprising taxa from Blendiella , Lobatotrema , Schistorchis and Sphincteristomum ( Fig. 5 View FIGURE 5 ). Sequences of S. carneus were gathered in a strongly supported clade (NJ=97). The Australian forms of S. carneus ex. A. stellatus were resolved into a well-supported clade (NJ=76) as a sister to the Egyptian form, S. carneus ex. A. hispidus , which in turn, is a sister to the other Australian form, S. carneus ex. A. manilensis .

Remarks. According to the established species recognition criteria proposed by Bray et al. (2022) and employed by Magro et al. (2023), combined molecular and host differentiation provides a solid foundation for the recognition of separate species. However, Magro et al. (2023) did not differentiate between specimens of S. carneus ex. A. stellatus and S. carneus ex. A. manilensis despite molecular distinctions and host differentiation. Some qualitative concerns were noted. First, they observed that the host-associated distinction in the molecular characterizations of these forms was somewhat confusing. Specifically, the low cox 1 molecular distinction between the two Australian groups of S. carneus (48 bp) compared to that observed between the most similar combination of Blendiella species (56–57 bp). Additionally, the ITS2 rDNA distinction (3 bp) between the two Australian forms of S. carneus mirrors that between the two species of Blendiella and exceeds the intraspecific variation seen in the two populations of Paraschistorchis stenosoma ( Hanson, 1953) Blend, Karar & Dronen, 2017 . Moreover, while the two Blendiella species differ by 8 bp in their 28S rDNA sequences, the Australian forms of S. carneus are identical in this region. Second, there is an absence of consistent morphological differences between specimens from A. manilensis and A. stellatus , with overlapping morphometric characteristics. Third, the marginal host distinction in S. carneus , due to the identification of additional hosts ( A. hispidus and L. lunaris ) harboring this species, raises doubts about the significance of host specificity or suggests a greater level of unrecognized diversity.

The ITS2 rDNA distinction of 3 bp between the two Australian forms of S. carneus is higher than that observed between the two entities of Lobatotrema aniferum Manter, 1963 (sensu lato) ( Table S3 View TABLE 3 ). Species delimitation analyses using cox 1 mtDNA indicated that the Australian records of L. aniferum represent two species with 21 bp differences in cox 1 mtDNA (Supplementary data; Table S4). Such differences are lower than that observed between specimens of S. carneus ex. A. stellatus and S. carneus ex. A. manilensis . Hence, the molecular distinction between the two entities of the Australian S. carneus is not lower than the level typically observed between combinations of species in schistorchiines.

Generally, phylogenetic trees using 28S rDNA, ITS2 rDNA, and cox 1 mtDNA revealed restriction of each schistorchiine species to one host type or a limited range of environmentally/taxonomically associated hosts ( Figs. 4 View FIGURE 4 , 5 View FIGURE 5 & S 1 View FIGURES 1 & 2 ). Species delimitation analyses referred to the presence of two distinct entities within the populations of both S. carneus and L. aniferum (Supplementary data; Fig. S1 View FIGURES 1 & 2 ). Each entity has been collected from a different host, suggesting that there may be even greater/unrecognized diversity within the genera of the Schistorchiinae and highlighting the significant role of host specificity. Additionally, the Australian records of P. stenosoma , collected from the large-scaled leatherjacket, Cantheschenia grandisquamis Hutchins, 1977 ( Tetraodontiformes : Monacanthidae ) at nearby localities (Lizard Island vs Heron Island), also diverged into two entities despite being from the same host (see Supplementary data; Fig. S1 View FIGURES 1 & 2 ). This reflects the potential existence of cryptic species, pointing to a higher unrecognized parasite richness in the Schistorchiinae .

The investigation Magro et al. (2023) referred to (i) resurrection of Lobatotrema as a ‘cryptic genus’ with Sphinteristomum where specimens of L. aniferum exhibited a phylogenetic divergence/separation from, but noticeable morphological similarity to, Sphincteristomum acollum Oshmarin, Mamaev & Parukhin, 1961 ; (ii) limitation of all reports of S. acollum on record from A. stellatus only, excluding records of Yamaguti (1971), in part, and Machida & Kuramochi (1999); and (iii) reassignment of the Japanese records of S. acollum ( Yamaguti 1971; Machida & Kuramochi 1999) as additional records of L. aniferum . Accordingly, we found that schistorchiine species exhibit a strong tendency towards high host specificity rather than species richness, indicative of the critical importance of host recognition in schistorchiines diversification, even those species difficult to differentiate morphologically.

To the best of our knowledge, few reports included records of two species from the Schistorchiinae from the same host. Typically, both belonged to the same genus and were very similar morphologically. Blend et al. (2017) collected Plesioschistorchis callyodontis (Yamaguti, 1942) Blend, Karar & Dronen, 2017 and Plesioschistorchis haridis ( Nagaty, 1957) Blend, Karar & Dronen, 2017 from the common parrotfish, Scarus psittacus Forsskål ( Perciformes : Scaridae ) from nearby localities in the northern Red Sea proximate to Hurghada, Egypt. The sandwich isle file, Cantherhines sandwichiensis (Quoy & Gaimard) ( Tetraodontiformes : Monacanthidae ) has been reported to harbor P. stenosoma and Paraschistorchis zancli ( Hanson, 1953) Blend, Karar & Dronen, 2017 from off the Hawaiian Islands ( Yamaguti 1970). Magro et al. (2023) reported Paraschistorchis seychellesiensis ( Toman, 1989) Blend, Karar & Dronen, 2017 from the honeycomb filefish, Cantherhines pardalis (Rüppell) ( Tetraodontiformes : Monacanthidae ), off Heron and Lizard Islands, Australia. Additionally, Skrjabin (1959), Pritchard (1963) and Yamaguti (1970) recorded P. stenosoma from the same host off Hawai. Such records indicate the possible presence of morphologically closely related schistorchiine species within the same host; and an apparent common phenomenon among members of this subfamily. This is also consistent with our observation of dividing records of P. stenosoma into two entities (i.e., two separate species) despite being collected from the same host but from two localities (Supplementary data; Fig. S1 View FIGURES 1 & 2 ). Additionally, through a comprehensive review of records of schistorchiines, almost all recognized species exhibited notable host specificity or parasitized a few closely related hosts except for L. aniferum , P. stenosoma and S. carneus . This raises question about the extent and possibility of species complexes as suggested by Magro et al. (2023) and clarified in the present study. Accordingly, we find host specificity among species of the Schistorchiinae to be of critical, not marginal, importance in providing a basis for recognition of species even those difficult to distinguish morphologically.

Genetic distances among the Schistorchiinae within the 28S and ITS2 rDNA regions are typically low; either a few nucleotides or none at all. In contrast, the cox 1 mtDNA gene shows notable nucleotide differences among closely related species, even when they are morphologically similar, and to a greater degree than can be considered intraspecific variation. Accordingly, we concluded: (i) insufficiency of the 28S and ITS2 rDNA regions for differentiating between closely related species, and low variation compared to cox 1 mtDNA; (ii) tendency of schistorchiines in ribosomal DNA region-based phylogenetic trees to follow the pattern of distribution correlated with oral sucker morphology whereas mitochondrial gene-based phylogenetic tree exhibit separations of taxa correlated with the type of host; (iii) variation in host species/genus is accompanied predominantly by a distinct taxon of schistorchiines, particularly with differences in host species/genus; and (iv) the lowest variations among 28S and ITS2 rDNA sequences of closely related taxa may indicate that they belong to different species even if there are strong morphological affinity.

We have noted some concerns regarding the robustness of the dataset used by Magro et al. (2023) to distinguish the two Austuralian populations of S. carneus , mainly due to dependence on only two specimens from a single A. manilensis (see Magro et al. 2023). Their low number of specimens is insufficient to determine the full range of variability of morphological measurements, and it cannot clarify whether intraspecific changes exist or not, especially given the inability to distinguish morphologically between the two populations. However, the strong molecular evidence demonstrating high nucleotide differences within the cox 1 mtDNA gene coupled with strongly supported species delimitation analyses based on molecular phylogeny shows that these two populations are genetically different from each other confirming the existence of a species complex of Australian populations of S. carneus separable by host—either A. stellatus or A. manilensis .

Cluster analysis

( Fig. 6 View FIGURE 6 ; Table 3 View TABLE 3 )

Our dendrogram divided records of S. carneus into two large groups, each one split into two clusters. These clusters were arranged from the closest to the more distant record in the following order: Cluster (A) consisted of specimens of Madhavi et al. (1986) from L. lunaris and records of S. carneus from A. stellatus ( Lühe 1906) and that from A. hispidus ( Blend et al. 2017; Present study); Cluster (B) consisted of specimens of Magro et al. (2023) from A. stellatus off Lizard Island; Cluster (C) was the largest one and was comprised of a group of records from Magro et al. (2023) from A. manilensis and A. stellatus , and records of S. carneus from A. stellatus ( Lühe 1906) and from A. hispidus ( Blend et al. 2017; Present study); and Cluster (D) which represented collections of S. carneus from A. hispidus ( Hafeezullah 1981; Blend et al. 2017; Present study). It was clear from Cluster analysis that i) Cluster (A) and Cluster (D) were the most distant groups; ii) specimens of Hafeezullah (1981) were the most isolated compared to other records of Cluster (D); and iii) generally, records of Lühe (1906), Blend et al. (2017) and present study were the most diverse and are distributed adjacent to each other among the other records in a consistent pattern, which reflects the high extent of diversity that these three records showed compared to the other records in addition to their high morphometric convergence.

Remarks. The wide distribution of records of Lühe (1906), Blend et al. (2017), and the present study in the cluster analysis using the Hierarchical agglomerative clustering reflected a high extent of diversity shown by those three records compared to the other records and refers to their high morphometric convergence. Although the three records were from two different hosts ( A. stellatus vs A. hispidus ) and distinctly distant localities (Sri-Lanka vs Egypt), their close morphometry could be attributed to the fact that (i) several common measurements were highly variable and apparently correlated to degree of worm maturation; (ii) both A. stellatus and A. hispidus are spread over a wide geographical range, and resultantly, so would be their parasites; iii) both A. stellatus and A. hispidus live in marine to brackish, reef-associated, tropical waters, distributed over a great portion of the Indo-Pacific, and they have nearly the same biology, food/feeding habitats, and co-exist in the same environment ( Froese & Pauly 2024). Hence, the possibility of transmitting the same parasitic taxon within both hosts is acceptable. The existence of S. carneus in two very closely associated hosts did not contradict host specificity but supported it, as it showed a clear specialization towards two ecologically and biologically close species. An example of this is the reporting of P. stenosoma from two closely associated hosts belonging to two monacanthid genera, Cantherhines pardalis and Cantheschenia grandisquamis ( Magro et al. 2023) .

The morphometric distinction of the specimens of Madhavi et al. (1986) from L. lunaris are in its own cluster away from other records of S. carneus and can be attributed to their critical morphometric ratios in contrast to all other records previously mentioned (i.e., morphometric measurements with low CV values) (see Results; Table 3 View TABLE 3 ). The extreme separation of the specimens of Madhavi et al. (1986) in the cluster analysis is consistent with the following morphological observations; (i) possessing an elongated and pear-shaped small seminal vesicle that extends posterior to the ventral sucker from the left side and narrows anteriorly to form a short pars prostatica vs a large, wide, oblong to clavate pouch possessing a transversally, sub-spherical to elliptical seminal vesicle, extending further dextral to the midline; (ii) presence of furrows in the pre-oral lobe region compared to other forms of S. carneus ; (iii) L. lunaris is demersal, oceanodromous, distributed in the Indo-West Pacific, occasionally enters estuaries, and occurs over sandy bottoms ( Froese & Pauly 2024) which is a different ecology from that of A. hispidus and A. stellatus , thereby casting doubt on the possibility of L. lunaris becoming infected with the same parasites that are present in A. hispidus and A. stellatus . Another supporting point is the question of Madhavi & Bray (2018) about marked differences in the body and egg sizes between the specimens of Hafeezullah (19861) and Madhavi et al. (1986). Specimens of Hafeezullah (1981) were much smaller in body dimensions (5,596 –6,000 µm × 2,076 –3,052 µm) and egg size (56–59 µm × 38–42 µm) than those of Madhavi et al. (1986) (12,100 –14,400 µm × 4,580 –4,800 µm; 78–80 µm × 28–32 µm).According to the aforementioned reasons, the restriction of schistorchiines to parasitizing a specific host or closely associated hosts, combined with other morphological differences observed in the specimens of Madhavi et al. (1986) (i.e., having anteriormost extent of the vitellarium restricted to level of either the ovary or the seminal vesicle, a dextro-submedian ovary and a hermaphroditic duct of equal length to the pars prostatica), we concluded that the specimens of Madhavi et al. (1986) likely represent a cryptic species and, thus, in need of future formal description.

The extreme position of the specimens of Hafeezullah (1981) in Cluster (D) is controversial. However, it can be attributed to measurements taken from two shrunken specimens as stated by Hafeezullah (1981), which resultantly, did not show the true extent of morphological, allometric and developmental variability. Regardless of measurements associated with worm maturity and growth such as the distance between suckers, we found that morphological features and morphometric ratios were consistent with Sri-Lankan and Egyptian specimens of S. carneus (see Table 3 View TABLE 3 ).

Within Cluster (C), the Australian population of S. carneus ex. A. manilensis gathered in its own clade as a sister to a group composed of Australian forms of S. carneus ex. A. stellatus off Moreton Bay with parts of Sri-Lankan form of Lühe (1906) and Egyptian forms. This convergence between the two clades is interpreted due to the close values of measurements only, however, insufficient to justify their identification as the same entity, S. carneus . This argument is supported by two main points. (i) Arothron manilensis inhabits marine muddy substrates and shallow estuaries and is restricted to the Western Pacific only whereas A. hispidus and A. stellatus inhabit outer reef slopes and are widely distributed in the Indo-Pacific and Eastern Pacific ( Froese & Pauly 2024). This variability among such environments despite their proximity, might be a factor influencing the potential differentiation of their associated parasites. (ii) High distinct molecular differences in the cox 1 mtDNA region between the Australian population of S. carneus vs A. stellatus and A. manilensis coupled with a molecular phylogeny-based species delimitation analysis separating S. carneus ex. A. manilensis from S. carneus ex. A. stellatus . Accordingly, we consider the Australian population of S. carneus ex. A. manilensis to be a hidden species, closely related to S. carneus , and in need of future formal description.

Specimens of Magro et al. (2023) from A. stellatus off Lizard Island clustered in its own clade as a basal to all members of the Clusters (C) and (D), and their separation from S. carneus ex. A. stellatus off Moreton Bay is explained based on several morphometric values in contrast to all other records (see Results above; Table 3 View TABLE 3 ). The consistency of all genetic markers between the specimens of Magro et al. (2023) from A. stellatus off Lizard Island with that off Moreton Bay are indicative that both Australian records belong to the same species. Measurements of both Australian groups were based only on two specimens of S. carneus from each locality, which did not allow for observation of the whole range of morphological variability. The inclusion of measurements of the two Australian specimens of S. carneus ex. A. stellatus as one population based on identical molecular data showed a wide range of measurements and morphometric ratios fell within, and were similar to, that observed in Sri-Lankan and Egyptian forms. Based on the forementioned reasons, we suggest Australian forms of S. carneus ex. A. stellatus are the same as that Sri-Lankan and Egyptian forms.