Morphological and molecular (28S rRNA) data of monogeneans (Platyhelminthes) infecting the gill lamellae of marine fishes in the Campeche Bank, southwest Gulf of Mexico

Abstract During the examination of 913 fish specimens belonging to four families in the Campeche Bank (Gulf of Mexico), 23 gill ectoparasitic monogenean species were found, which belong to three families: Dactylogyridae, Microcotylidae and Diclidophoridae. The species Euryhaliotremaamydrum, E.carbuncularium, E.dunlapae, E.fajeravilae, E.fastigatum, E.longibaculum, E.paracanthi, E.tubocirrus, Haliotrematoidescornigerum, H.gracilihamus, H.heteracantha, H.longihamus, H.magnigastrohamus, H.striatohamus, Hamatopedunculariabagre, Neotetraonchusbravohollisae, and N.felis (all Dactylogyridae) were found on the hosts Lutjanussynagris, L.griseus, Ariopsisfelis, Bagremarinus, Archosargusrhomboidalis, and Haemulonplumieri. Additionally, Microcotylearchosargi, Microcotyle sp., and Microcotyloidesincisa (all Microcotylidae) were found on L.griseus and A.rhomboidalis; finally, Choricotyle sp. 1, Choricotyle sp. 2, and Choricotyle sp. 3 (all Diclidophoridae) were found on H.plumieri. The prevalence, abundance, mean intensity of infection, and supplementary taxonomic revisions for all monogeneans found are provided. Partial sequences of the 28S rRNA gene were also obtained for monogeneans of ariid, sparid, and haemulid host fishes to explore their systematic position within the Monogenea. New locality and host records for some previously described species of Euryhaliotrema, Hamatopeduncularia, Microcotyle, and Choricotyle from lutjanid, ariid, sparid, and haemulid hosts were reported. The present study adds evidence supporting the interoceanic occurrence of the same monogenean species (on lutjanids) on the west-east Atlantic and Pacific Oceans (= amphiamerican species). As previously suggested, there are at least, two possibilities to explain that parasite distribution: differentiation of morphological features in these monogeneans have resulted in only slight to insignificant morphological changes developing over the extended period of 3.2 mya (when the Isthmus of Panama was closing) and/or speciation is only evident at molecular level.


Introduction
The Campeche Bank (southwest Gulf of Mexico) represents an important marine ecosystem characterized by a high biodiversity, which is threatened by important overfishing and energy (petroleum) extraction activities (Soto et al. 2014). For example, oil can affect marine wildlife by physical effects, i.e., death by suffocation, with oil blocking air passageways or fish gills (NOAA 2018, Overstreet andHawkins 2017). Because of its economic impact on Mexico´s economy, the Campeche Bank is considered a strategic region in the national plans for the social and economic development of Mexico (Piñeiro 2001).The knowledge of the diversity, abundance and distribution of species is the base for developing management plans for threatened species and preserving its natural resources for ecological and economic purposes (Ocean Conservancy 2011).
Biodiversity is widely considered to correlate with ecosystem health, the presence or abundance of parasites becomes part of that positive biodiversity. Otherwise, the fewer the parasites observed, the worse the environmental conditions and thus the biodiversity (see Vidal-Martínez and Wunderlich 2017). Therefore, parasite biodiversity information can be critical for the control and safe management of commercial fish species (Vignon andSasal 2010, Quiazon 2015). However, parasites remain an underestimated component of the total biodiversity in many regions (Lafferty et al. 2015).
Despite their ecological and environmental effects, there have been few studies aimed at collecting and examining fish samples for parasites; in consequence, many parasite species go undetected or are poorly studied. Low availability and poor quality of material for examination also adds to this problem. This lack of knowledge about biodiversity also prevents us for understanding the connectivity between the northern and southern Gulf fisheries.
As part of a research project on fish parasite biodiversity in the Campeche Bank, we had the opportunity to undertake a survey of ectoparasitic monogeneans infecting the gill lamellae of six marine fish species. Here, we provided: 1) supplementary information and illustrations of the sclerotised and/or soft structures of the monogenean species found; 2) information on the prevalence and intensity of infections at each site sampled; and 3) data on the biometrical variability of individual monogenean species collected on different hosts. In addition, partial sequences of the 28S rRNA gene (D1-D3) were amplified from monogeneans of ariid, sparid, and haemulid hosts to explore their systematic position within the Monogenea. The occurrence of the monogenean species found with respect to the west-east Atlantic and Pacific divide is briefly discussed.

Materials and methods
We studied the most abundant fish species (Diario Oficial de la Federación 2012), resulting in six species that were caught from three coastal locations in the state of Campeche [southwestern coast of the Gulf of Mexico: San Francisco (19°55.988'N;90°41.969'W),Seyba Playa (19°42.580'N;90°44.155'W),and Champoton (19°16.390'N,90°49.194'W)]. Fish were collected over a period of eight months (from January to August 2015) using gill nets. Fish were kept on ice for a maximum of 12 hours before their gills were removed and placed in fingers bowls containing a 4% formaldehyde solution to fix ectoparasites. Parasites were subsequently detached from the gills using fine needles under a dissecting microscope, stained with Gomori's trichrome stain and mounted in Canada balsam (Vidal-Martínez et al. 2001). A selection of specimens was mounted on slides using a mixture of lactic acid (LA) and glycerin-ammonium picrate (GAP) and then remounted in Canada balsam (Mendoza-Franco et al. 2013) to obtain measurements of the haptoral structures and copulatory complex.
Prior to DNA analysis, parasites were fixed with 96% ethanol and individually identified based on the morphology of their haptors. The haptor of each specimen was removed using syringe needles (used for insulin injections) and mounted unstained in a mixture of LA and GAP. The body of the worm was transferred to a labeled Eppendorf tube containing 96 % ethanol and stored at room temperature until required for molecular evaluation. Processed haptors were then remounted in Canada balsam (see Mendoza-Franco et al. 2009) and studied using an immersion oil objective on a DM2500 Leica microscope. These haptors were kept as molecular vouchers (hologenophore, i.e., the voucher specimen from which the molecular sample was obtained; see Astrin et al. 2013) and deposited in the CNHE.
Two to ten bodies of excised specimens from the gills of ariid, sparid, and haemulid fishes collected at each of the three sampling sites were placed individually in a 0.2 µl Eppendorf tube for genomic DNA extraction. Genomic DNA of each individual was extracted using 20 µl Chelex (100 sodium) and 5 µl proteinase K (at 10mg µl -1 ) to lyse parasite tissues. Specimens were immediately incubated for 3 h and 15 minutes at 96 °C to denature the proteinase K. Volumes of 5 µl were taken from each lysed preparation to serve as template DNA samples in the PCR assays. A fragment of the 28S rRNA gene (D1-D3) was amplified using the polymerase chain reaction (PCR). The internal primers Halio-F (5´-ACCCGCTGAATTTAAGCAT-3´) and Halio-R (5´-TGGTCCGTGTTTCAAGAC-3´) were used for amplification (García-Vásquez et al. 2015). All PCR reactions were performed in a final volume of 50 µl composed of 5 µl 10× PCR buffer, 1.5 µl 10 mM dNTPs mixture (10 µM each), 4.0 µl 2.0mM MgCl 2 , 1.5µl of each primer (10 µM), 5 µl template DNA, 0.24 µl Taq DNA polymerase (1.2 units), and 31.26 µl of sterile distilled water. The following thermocycling profile was used: initial denaturation at 94 °C for 2 min, followed by 35 cycles of 94 °C for 30 sec, annealing at 55 °C for 30 sec and final extension at 72 °C for 3 min. The mounts or permanent preparations containing a haptor used to identify parasite specimens for which the body was used to amplify DNA were deposited in the CNHE.

Alignment, phylogenetic analyses, and sequence divergence
28S (D1-D3) sequences obtained in the current study were aligned with that of other monogenean species available in GenBank using Muscle algorithm implemented in Mega 7 (Kumar et al. 2015) and adjusted manually with the program Mesquite 2.75 (Maddison and Maddison 2011). The software jModelTest version 2.1.10 (Darriba et al. 2012) was used to select the best model of evolution for our dataset. The model (GTR+I+G) was selected based on the Akaike information criteria. Maximum likelihood (ML; 1000 Bootstrap) and Bayesian Inference (BI) analyses were performed using Mega 7 and Mr. Bayes version 3.2, respectively (Huelsenbeck and Ronquist 2001). Mr. Bayes was used based on Markov chains model with burning periods every 1,000 generations to reach a consensus after 400,000 generations. Numbers at the interior branches of the consensus tree represent posterior probabilities (PP) and booptstrap of maximum likehoods. Trees were drawn using the program Fig Tree V .1.4.3 (Drummond et al. 2006). The genetic divergence among species [Hamatopeduncularia bagre Hargis, 1955, Microcotyle archosargi MacCallum, 1913, Haliotrematoides striatohamus (Zhukov, 1981 Mendoza-Franco, Reyes-Lizama & González-Solís, 2009, and Choricotyle spp.] was estimated using the uncorrected "p-distances" method with the program MEGA v. 5 (Tamura et al. 2011).

Euryhaliotrema Kritsky & Boeger, 2002
Euryhaliotrema Kritsky & Boeger, 2002: 12, fig. 1 Comments. This species was originally described on the sheepshead A. probatocephalus from the Indian River Lagoon in Florida (Kritsky and Bakenhaster 2011). This species is mainly characterized in having a tightly coiled MCO and dorsal anchor roots approaching the length of the dorsal anchor shaft. Differences in the length of the dorsal anchors between present specimens and those of E. amydrum originally described were found (i.e. length 40-45 vs. 49 53), but the worms are clearly conspecific. Montoya-Mendoza et al. (2015) reported E. amydrum on A. probatocephalus from Alvarado Lagoon and El Conchal estuary in Veracruz (Gulf of Mexico). However, these latter authors did not provided any accession number for their parasite specimens apparently deposited in the CNHE. Then, we could not corroborate Montoya´s finding.
Molecular data. The present study also provided the first molecular data of E. carbuncularium; there are two sequences (676 and 856 bp, respectively) of individual specimens of this monogenean species included within the analyses that shows that this species forms a sister lineage to that containing Euryhaliotrema mehen (Soler-Jiménez, García-Gasca & Fajer-Ávila, 2012) Kritsky, 2012, which is known on Lutjanus guttatus (Steindachner, 1869) in the Eastern Pacific (see Figure 1).
Specimens deposited. Three reference specimens in the CNHE (10608). Two slides, each containing a haptor of a specimen of E. carbuncularium used to amplify its DNA are deposited in the CNHE (10622).
Representative DNA sequence. GenBank accession number MG586874, MG586875.  (Kritsky and Bakenhaster 2011). Morphometrical comparison between present specimens and those originally described did not reveal a significant difference. As mentioned above for E. amydrum, Montoya et al. (2015) also reported E. dunlapae on A. probatocephalus in Veracruz, Mexico; however, they did not provide any accession number for these specimens of E. dunlapae.

Euryhaliotrema dunlapae
Specimens deposited. Seven reference specimens in the CNHE (10609). Comments. This species was described from L. argentiventris from the Perlas Archipielago, Panama by Kritsky and Mendoza-Franco (in Kritsky 2012). Euryhaliotrema fajeravilae is distinguished from other species of the genus infecting lutjanids by having larger anchors and a noticeably smaller copulatory complex. The morphometrics of the present specimens did not differ from that of the original description.

Haliotrema fastigatum Zhukov, 1976
Haliotrema fastigatum Zhukov, 1976: 43, fig. 10 Comments. Zhukov (1976) originally described this species as Haliotrema fastigatum from L. apodus and Lutjanus jocu (Bloch & Schneider, 1801) from the area Havana (Gulf of Mexico). In 2002, Kritsky and Boeger transferred this species to Euryhaliotrema as E. fastigatum based on details presented in the original description (Zhukov 1976) of the copulatory complex, internal organs, and haptoral armament according with the diagnosis of Euryhaliotrema. Later, Kritsky (2012) redescribed E. fastigatum based on specimens collected from L. griseus and other lutjanids (L. apodus, L. jocu, and L. argentiventris) from Florida and off Taboga Island, Perlas Archipielago, Isla Saboga, and Isla Tabugilla (all from the Pacific of Panama). Euryhaliotrema fastigatum is characterized in having a thinning of the base of the dorsal anchor near its junction with the anchor shaft and by lacking by lacking an articulation process in the copulatory complex. Measurements and the morphology of the sclerotized structures of the present specimens do not differ significantly from that figured in the redescription of E. fastigatum. This monogenean species has also been reported on Lutjanus analis (Cuvier, 1828) and L. griseus from Puerto Rico and off Venezuela (Bosques-Rodriguez 2004, Fuentes Zambrano et al. 2003, Fuentes Zambrano and Silva Rojas 2006, Kritsky 2012. Specimens deposited. Ten reference specimens in the CNHE (10621).

Haliotrema longibaculum Zhukov, 1976
Haliotrema longibaculum Zhukov, 1976 Comments. Euryhaliotrema longibaculum was originally described and depicted (as Haliotrema longibaculum) from L. synagris and L. mahogoni collected off Cuba (Area Havana) (Zhukov 1976, Kritsky 2012. Later, Kritsky and Boeger (2002) transferred the species to Euryhaliotrema based on Zhukov´s original description and drawings. The present specimens fit the diagnosis of E. longibaculum, which was redescribed by Kritsky (2012) based on specimens found in L. synagris from Florida, USA. Euryhaliotrema longibaculum is characterized by having dorsal anchors with an elongate superficial root, poorly developed deep root and elongate point extending anteriorly near to the level of the union of the anchor shaft and base, and an articulation process in the copulatory complex connecting the accessory piece to the base of the MCO. Morphometrical comparison of the present material with the redescription of this species provided by Kritsky (2012) did not reveal any differences. Recently, Montoya et al. (2016) reported E. longibaculum (voucher CNHE 10221), from L. synagris from Santiaguillo Reef, Veracruz (Gulf of Mexico). Examination of that voucher allowed us to confirm the species identity.

Haliotrema paracanthi Zhukov, 1976
Haliotrema paracanthi Zhukov, 1976: 42-43 Comments. This species was originally described as Haliotrema paracanthi by Zhukov (1976) from L. apodus from the Area Havana (off Cuba) and later transferred to Euryhaliotrema by Kritsky and Boeger (2002) based on the drawings presented in the original description of Zhukov (1976). Subsequently, Kritsky (2012) redescribed E. paracanthi based on specimens collected from L. jocu and other lutjanids (L. argentiventris and L. griseus) from Florida and off Taboga Island, and Perlas Archipielago (both from Panama). Euryhaliotrema paracanthi is differentiated from other species of Euryhaliotrema infecting lutjanids by possessing a subterminal spine or hook on the accessory piece. The morphometrics of the present specimens did not differ from those reported in the redescription of E. paracanthi.

Haliotrema tubocirrus Zhukov, 1976
Haliotrema tubocirrus Zhukov, 1976 Comments. This species was originally described as Haliotrema tubocirrus from the gills of L. synagris, L. analis, L. apodus, Lutjanus cyanopterus (Cuvier, 1828) and Rhomboplites aurorubens (Cuvier, 1829) from the environs of Havana, Cuba (Zhukov 1976). Kritsky and Boeger (2002) transferred this species to Euryhaliotrema based on the description and drawings provided in the original description by Zhukov (1976 Comments. Identification of present specimens is based on diagnosis provided by Kritsky et al. (2009b) which transferred this species from Haliotrema to Haliotrematoides on the basis of study of Zhukov's (1976) original figures and those provided by Bosques-Rodríguez (2004). This species is characterized in having an inner spur on the dorsal anchor shaft and shaft of MCO having a proximal loop (see Kritsky et al. 2009). Haliotrematoides cornigerum is currently reported from L. synagris and L. mahogoni from the Bay of Campeche (Area Havana) and Puerto Rico (Zhukov 1976, Bosques-Rodríguez 2004, Kritsky et al. 2009b Comments. This species was originally described as Haliotrema gracilihamus on L. apodus from Campeche Bay (Area Havana) (Zhukov 1976 (Kritsky et al. 2009b). This species is differentiated from its congeners in having a coiled tube of the MCO with two complete counterclockwise rings and ventral bar with posteromedial shield-like process, and anteromedial flap having two bilateral pockets. Morphometric comparison of the present material with the redescription of this species provided by Kritsky et al. (2009b) did not reveal any differences.
Specimens deposited. Ten reference specimens in the CNHE (10606). Figure 2 Haliotrema heteracantha Zhukov, 1976: 36-37, fig. 3 Comments. This species was originally described as Haliotrema heteracantha from L. synagris by Zhukov (1976) who also reported it from other five lutjanids [L. mahogoni, L. apodus, Ocyurus chrysurus (Bloch, 1791), L. analis, and L. griseus] from Bay of Campeche (Area Havana) (Zhukov 1976). Subsequently, Kritsky et al. (2009b) transferred this monogenean species to Haliotrematoides by based on original figures of this species made by Zhukov (1976). It has been stated that Hal. heteracantha shows a notable similarity with Hal. guttati in the Pacific coast off Mazatlán, Sinaloa Mexico based on the comparative morphology of the anchors (i.e. dorsal and ventral anchors with spurs on the inner surfaces of the anchor shafts), bars, and copulatory complex (see Kritsky et al. 2009b).

Haliotrema heteracantha Zhukov, 1976
Both monogenean species are currently considered distinct based on the absence of a loop in the shaft of the MCO in H. heteracantha (present in Hal. guttati). However, examination of present specimens of H. heteracantha showed that morphology of the MCO is variable and a loop is present as well in the shaft of the MCO (see Figure 2). Accordingly, it would suggest that H. guttati is a junior synonym of H. heteracantha. However, the two species have been isolated since formation of the Panamanian Isthmus (~ 3 mya), which theoretically it would support they are distinct species. Sequences of both could probably help in answering the question of conspecificity. Montoya et al. (2016) reported E. heteracantha (voucher CNHE 10218) from L. synagris from Santiaguillo Reef, Veracruz (Gulf of Mexico). Examination of that voucher allowed us to confirm the species identity.

Haliotrema magnigastrohamus Zhukov, 1976
Measurements of the present finding fits well with the morphometric of H. magnigastrohamus provided by Kritsky et al. (2009b). Montoya et al. (2016) reported H. magnigastrohamus on L. synagris from Santiaguillo Reef, Veracruz (Gulf of Mexico) and deposited a voucher specimen in the CNHE (accession number 10220). However, examination of that specimen revealed it to be an Euryhaliotrema sp. that resembles E. torquecirrus. Examination of another voucher specimen labeled as E. torquecirrus (CNHE 10223) on L. synagris deposited by the same authors revealed it to be same form as that of Euryhaliotrema sp. In this latter form, the coil of the MCO comprises 2½ rings (more than four rings in E. torquecirrus) and a single accessory piece (accessory piece includes two components in E. torquecirrus) (see Euryhaliotrema sp. in Figure 4 and E. torquecirrus in figure 24 in Kritsty 2012).

Haliotrema striatohamus Zhukov, 1981
Haliotrema striatohamus Zhukov, 1981: 179, fig. 1 Figure 1) from L. guttatus off Taboga Island (type locality) and Perlas Archipielago in Pacific waters of Panama (Kritsky et al. 2009b  Comments. Hamatopeduncularia bagre was originally described on B. marinus from Alligator Harbor, Franklin County, Florida, USA (Hargis 1955a). Recently, this species was redescribed based on specimens found on another catfish, Bagre bagre (Linnaeus, 1766) from several localities in Brazil (Domingues et al. 2016). This monogenean species is characterized mainly by the possession of hooks on haptoral digits, double dorsal bar, and dissimilarity in the size of anchors. Measurements and the morphology of the sclerotized structures of the present specimens do not differ significantly from that figured in the redescription of H. bagre. Molecular data. A 768-770 bp fragment of the 28S rRNA gene (D1-D3) of H. bagre on B. marinus was obtained in the present study, which represents the first molecular data for this monogenean. There are three sequences of individual specimens of H. bagre included into the analyses that revealed that this species forms a sister lineage to that containing N. felis (see Figure 1). Comments. In A. felis a simultaneous infection with N. felis was found. Since all worms could not be identified, the data on infection rate relate to N. bravohollisae and N. felis. Neotetraonchus bravohollisae was originally described on Galeichthys felis (Linnaeus) (now A. felis), from Dauphin Island, Alabama coast, Gulf of Mexico (Paperna 1977), and posteriorly reported on Hexanemathichthys assimilis [now Ariopsis assimilis (Günther, 1864)] from Chetumal Bay, Yucatan, Peninsula on the border between Mexico and Belize (Aguirre-Macedo et al. 2007). More recently, N. bravohollisae was redescribed based on its type specimens and other specimens collected on A. felis in the Gulf of Mexico off the Yucatan, Peninsula Gulf of Mexico (Telchac Puerto and Port of Celestun) (Kritsky et al. 2009a). Measurements and the morphology of the sclerotized structures of the present specimens fit well with those redescribed by these latter authors.

Ancyrocephalus felis Hargis, 1955
Ancyrocephalus felis Hargis, 1955a: 186-187, figs 28-33;Yamaguti 1963 Comments. This species was originally described as Ancyrocephalus felis on G. felis (now A. felis) from Alligator Harbor, Franklin County, Florida (Hargis 1955a). Yamaguti (1963) transferred this monogenean species to Haliotrema as H. felis based on the original description and his observations of the type specimens. Paperna (1977) transferred it to Neotetraonchus as N. felis and added a character within genus, the presence of an accessory piece in the copulatory complex and the onchium (accessory bar) in the haptor (see Kritsky et al. 2009a). Recently, N. felis was redescribed based on examination of its type specimen and other new specimens collected on A. felis from the Gulf of Mexico off Mississippi and the Yucatan Peninsula (Kritsky et al. 2009a).
Present specimens are clearly conspecific with those of N. felis from A. felis as redescribed by these latter authors. All these specimens have an elongate tube of the MCO directed posteriorly and reaching level of anterior end of germarium. Currently, N. felis has been reported on A. felis from Dauphin Island, Alabama coast, Gulf of Mexico (Paperna, 1977); West Ship Island, northern Gulf of Mexico off Mississippi, USA; Gulf of Mexico off Telchac Puerto and Port of Celestun, Yucatan, Mexico (Kritsky et al. 2009a). Present study also provided the first molecular data of N. felis by amplifying an 772 bp fragment of the 28S rRNA gene (D1-D3). There is one sequence of an individual specimen of N. felis included into the analyses that revealed that this species forms a sister lineage to that containing H. bagre occurring on other ariids, B. marinus and A. felis (see Figure 1). Specimens deposited. Four reference specimens in the CNHE (10616). Another slide containing a haptor of a specimen of N. felis used to amplify its DNA is deposited in the CNHE (10801).
Supplementary observations (measurements based on nine specimens) in Table 2. Comments. Specific placement of current specimens are in agreement with diagnosis provided by MacCallum (1913) who described this species from A. probatocephalus obtained from a fish market (origin unknown) in New York City, USA. Caballero y Caballero and Bravo-Hollis (1972) erected Paramicrocotyle to describe P. tampicensis and P. atriobursata on Diapterus olisthostomus (Gerreidae) (now Diapterus auratus Ranzani, 1842) from Ciudad Madero, Tamaulipas (Gulf of Mexico) as well as accommodate whithin the genus other sixteen species previously placed in Microcotyle, including M. archosargi. However, all species of Paramicrocotyle were reassigned to Microcotyle by Mamaev (1986), who considered Paramicrocotyle a junior subjective synonym of Microcotyle. Currently, M. archosargi (sensu Mamaev 1986) has been recorded from sheepshead (as Archosargus oviceps) taken at Alligator Harbor, Florida, by Hargis (1956); Iruegas-Buentello (1999)  These latter authors stated that M. archosargi has two bilateral zones of small spines lying slightly posterior to the armed genital atrium, which are close to the ventral surface of the worm, but somewhat deeper within the body than those of the genital atrium. We fully concur in these morphological observations based on examination of present specimens (see Figure 5). Based on examination of five vouchers (CNHE 0323) of M. tampicensis (Caballero y Caballero & Bravo-Hollis, 1972), it shows to be extremely similar to the general characteristics of M. archosargi, particularly in having morphologically comparable genital atrium (see figures 7-12 in Caballero y Caballero and Bravo-Hollis 1972; figure D in MacCallum 1913; Figure 5 in the present study). The resemblance of both M. tampicensis and M. archosargi can be explained by the fact that the former was mainly described and/or differentiated of other congeneric species based on the structure and shape of the genital atrium. The two species are presently considered distinct based on the length of the genital atrium, i.e., 279 in M. tampicensis vs. 105-180 in M. archosargi (see Table 2).
However, the five vouchers of M. tampicensis were flattened and/or distorted (i.e., one specimen with distorted genital atrium, two specimens with incomplete haptor and another specimen was fragmented in three parts) due to coverslip pressure, which may have altered the length of the genital atrium. Determination of possible synonymy, therefore, will depend on recollection of the specimens of M. tampicensis from D. olisthostomus in the Gulf of Mexico for comparison with M. archosargi. In other features, present specimens of M. archosargi from A. rhomboidalis metrically fit within range from those specimens found on A. probatocephalus (see Table 2). Differences in the number of testes and clamps, morphologically identical in specimens of M. archosargi from different hosts and locations, are considered as intraspecific variation. Montoya et al. (2015) reported M. archosargi on A. probatocephalus in Veracruz, Mexico, without providing any accession reference number from the CNHE. Then, we could not corroborate finding of these latter authors.
Molecular data. This study also provided the first molecular data for M. archosargi by adding a sequence (   Paramicrocotyle atriobursata] in the general morphology of the genital atrium, 1) two bilateral zones of small spines lying posterior to the armed genital atrium; 2) two posterolateral cavities; 3) genital atrium elliptical, occupying all postbifurcal area; in ventral view, the anterior margin of the atrium is gently curved; posteriorly, the atrium expands into a triangular shape to form an internal cavity surrounded by ventral lips with spines; anterior margin is projected as an operculum on the posterior margin (present in current specimens and M. atriobursata) (see figures 1 and 5 in Caballero y Caballero and Bravo-Hollis, 1972; Figure 5 in the present study); 4) number of testes, i.e. 21-22 vs. 20-35 in M. archosargi (MacCallum, 1913) and 20-25 in M. atriobursata. Microcotyle sp. differs from these two microcotylids in the width of the genital atrium, i.e. 155-175 vs. 211-242 in M. atriobursata and 80 in M. archosargi. Although current specimens are clearly members of the Microcotyle they were unsatisfactory to clarify details of internal organs for species identification. While intraspecific variation between individuals of Microcotyle sp. and M. archosargi might represent a single species, new collections of specimens of the former species are necessary for completing formal specific identification of this species.
Comments. Placement of present specimens in Choricotyle is based on examination of original descriptions of other species allocated or currently assigned to the genus in Fujii (1944), Hargis (1955b), Kritsky and Bilqees (1973), Oliva (1987), Lamothe-Argumedo et al. (1998), Oliva et al. (2009), andCohen et al. (2011). While the eight specimens of Choricotyle sp. 1 were unsatisfactory to clarify details of internal organs for species identification, they appear to represent an undescribed species based on the general morphology of the haptor and genital atrium. Choricotyle sp. 1 resembles C. anisotremi Oliva, 1987 (Desmarest, 1823) from Tortugas, Florida USA. All these monogeneans share the following features: presence of a sucker on internal quadrant on clamp (present in Choricotyle sp. 1 and C. anisotremi), relatively similar morphometry of clamps (i.e., 150-362 long × 130-325 vs. 152-219 in diameter in C. aspinorcha), number of spines of the genital atrium (10 spines in Choricotyle sp. 1 and C. aspinorcha), and a lappet with one pair of hooks, each with 27-33 long (one pair, each with 28 long in C. hysteroncha). Choricotyle sp. 1 differs from these three latter monogenean species by number of testes (12 vs. 90 in C. anisotremi, 42-88 in C. aspinorcha, and 6-7 in C. hysteroncha).
The finding of Choricotyle sp. 1 constitutes the second record (the first being that of Choricotyle leonilavazquezae Lamothe-Argumedo, Aranda-Cruz & Pérez-Ponce de , that occurs on the Pacific coast of Mexico) of a species of Choricotyle in Mexico and the first record on H. plumieri. In the present study, three species of Choricotyle (i.e., Choricotyle sp. 1, Choricotyle sp. 2 and Choricotyle sp. 3) were identified on this latter host species (see below) based on morphological features of the genital atrium, clamps and hooks on terminal lappets (when present), if they actually represent different species since variability in these diclidophorids might exhibit intraspecific differences in the shape or size of these structures above mentioned (see Yang et al. 2007).
Molecular data. The present study also provided the first molecular data on species of Choricotyle in Mexico; both sequences of Choricotyle sp. 1 included into the present analyses revealed that this species forms a sister lineage to that containing C. anisotremi (see Figure 6) which occurs on A. scapularis (Pomadasyidae) from Chile (Oliva 1987).
Other two slides, each containing a haptor of a specimen of Choricotyle sp. 1 used to amplify its DNA are deposited in the CNHE (10624 and 10625).
Comments. Choricotyle sp. 2 has the characteristics and features of Choricotyle (i.e. species having four pairs of clamps and genital spines ranging from seven to twelve and exceptionally, from 28 to 30 in Choricotyle rohdei Cohen, Cardenas, Fernandes & Kohn, 2011). Choricotyle sp. 2 appears closest morphologically to Choricotyle sp. 1 based on the presence of concentric arcs of small skeletal rods in dorsal fields of clamp and terminal lappet with one pair of hooks of relatively similar size (i.e., 33-35 long vs. 27-33 in Choricotyle sp. 1) and having a filament connecting shank and base (see Figure 7F, G). Choricotyle sp. 2 differs from Choricotyle sp. 1 in the general morphology of clamps (subrectangular vs. rod-shaped posterior portion of the medial sclerite, see Mp in Figure 7C), hooks (robust vs. slender shanks, respectively, see Sh in Figure 7) on the terminal lappet, and number of spines of the genital atrium (9 vs. 10). Only two specimens of Choricotyle sp. 2 found on H. plumieri that were flattened and unstained with GAP precluded determination of internal anatomy and the consequent specific assignment of the specimens. A determination may be possible given a more extensive revision of specimens to formally describe this species.
Choricotyle sp. 3 Figure 7E Present study. from H. plumieri was immature (less than one mm in total length). Reproductive organs were minimally or undeveloped to determine the specific assignment precluding resolution of the specimen as new or previously described. However, assignment of the current specimen to Choricotyle is based on the morphological similarity of its haptoral sclerites to those of species of Choricotyle described above on H. plumieri.

Discussion
In this study, we identified 23 gill-infecting monogenean species, assigned to three families (Dactylogyridae, Microcotylidae and Diclidophoridae) and seven genera (Euryhaliotrema, Haliotrematoides, Hamatopeduncularia, Neotetraonchus, Microcotyle, Microcotyloides, and Choricotyle), on marine fishes belonging to four families (Lutjanidae, Ariidae, Sparidae, and Haemulidae) from the Campeche Bank (southwest Gulf of Mexico) (see Table 1). Most monogeneans found on lutjanids in the present study were originally described from the area Havana (Gulf of Mexico) by Zhukov (1976). However, it is not known if Zhukov obtained lutjanids [i.e., O. chrysurus, L. apodus, and L. analis (all reported as hosts of H. heteracantha, H. gracilihamus, E. paracanthi, E. fastigatum, and H. magnigastrohamus)] in fish markets or within a commercial fisheries landing site wherein all fishermen might have been working within a radius of the harbor/Havana or if he may have known that some boats were fishing in the Campeche Bank (or the whole Gulf ) but landing those Campeche fishes in the port of Havana. In any case, the present survey of monogeneans on lutjanids in the Campeche Bank could represent new locality records as stated in Table 1. In the Campeche Bank, E. fajeravilae on L. griseus is reported for the first time; this monogenean species along with E. fastigatum, E. paracanthi, Hal. gracilihamus, and M. incisa in the Gulf of Mexico have previously been described and/or reported from the Pacific (Kritsky 2012, Kritsky and Boeger 2002, Kritsky et al. 2009, Mendoza-Garfías and Pérez-Ponce de León 1998. The occurrence of geminate species pairs of Monogenea off North America (as those mentioned above) has been thought to have developed through a vicariant co-evolutionary model when the Panamanian Isthmus divided historical host and parasite distributions into eastern Pacific and western Atlantic populations about 3.2 mya (see Kritsky 2012). However, considering the amount of time that has passed since the closing of the isthmus, and that monogeneans from the two oceans are so close morphologically (i.e., the putative pair represented by Hal. heteracantha in the Gulf of Mexico and Hal. guttati in the Pacific; see Comments for Hal. heteracantha) as to preclude separation is an issue that remains unclear. In fact, some monogenean species ranging on both sides of the isthmus have been provisionally accepted as different species until the putative impact of the Panamanian Isthmus on speciation within this group of parasites is determined (see Kritsky 2012). These putative pairs could suggests that differentiation of morphological features in the Monogenea is a comparatively long process, which in the amphiamerican clades resulted in only slight to insignificant morphological changes developing over the extended period of 3.2 mya and/ or speciation is only evident at molecular level (Kritsky 2012). The point is that other monogenean species could have speciated independently on their respective hosts in both sides of the Isthmus.
Molecular data from the present study provides evidence supporting morphological speciation of other monogeneans occurring on both sides of Isthmus. For example, E. carbuncularium from A. rhomboidalis from Campeche (Gulf of Mexico) appears to be phylogenetically associated with E. mehen from L. guttatus in the Eastern Pacific. Similarly, Haliotrematoides striatohamus from Haemulon plumieri appears to be a sister species of the clade containing Haliotrematoides guttati and Haliotrematoides spinatus, both from L. guttatus in the Pacific waters of Panama (see Figure 4).
In summary, the present study provided six novel sequences of the 28S rRNA gene that advance our understanding of the morphology and host-parasite associations of other monogenean groups. For example, M. archosargi from the sparid A. rhomboidalis from Campeche clustered with other microcotylids (M. sebastis, M. erythrini and M. arripis) described and/or reported on perciform (Sparidae) and scorpaeniform (Sebastidae) fishes ( Figure 6). All these microcotylids exhibit little differentiation at the molecular level despite substantial morphological differentiation on their respective geographically distant host species. Thus, either the 28S rRNA gene is a highly conserved region in these microcotylids or these monogeneans represent same species. Sequences of mitochondrial DNA COI could allow a better phylogenetic resolution of these monogeneans. However, knowledge of potential genes to be amplified in these monogeneans is very poorly known, especially for marine tropical species.
Similarly, in some instances, congeneric and phylogenetically related monogeneans infecting hosts of the same family appear to be phylogenetically closely related based on 28S rRNA gene. For example, Choricotyle sp. 1 from the haemulid H. plumieri appears to be related to C. anisotremi on another haemulid, A. scapularis from Chile. Finally, sequences of H. bagre on B. marinus (present study) (also present on B. bagre from Brazil) show that this monogenean is a sister species of N. felis on A. felis (see Figure 1), with both monogeneans on their respective ariid catfishes occurring from the western Atlantic (i.e. Florida and off Mississippi USA, Gulf of Mexico, Telchac Puerto and Port of Celestun, Yucatan, Mexico and northern Brazil). The relationship observed between these monogeneans is also congruent with that revealed in the phylogeny of their ariids hosts using also molecular data (see Betancur 2009). For example, the clade containing B. bagre and B. marinus (hosts of H. bagre) represents a basal position and genetically distant to that containing A. felis (host of N. felis) (see Betancur 2009: fig.  A). Furthermore, B. bagre appears to be a sister species of B. marinus (Betancur 2009).
Bagre bagre and B. marinus share the same monogenean species, H. bagre, suggesting that this monogenean has coevolved with both ariid hosts since their divergence from a common ancestor or the same monogenean species was able to infect these two closelyrelated catfishes after they diverged which is not "coevolved" per se; it is simply a lack of host specificity among congeneric hosts.