Tytthosoceros, Litvaitis & Bolaños & Quiroga, 2019

Pseudocerotidae Lang, 1884 View in CoL

A monophyletic Pseudocerotidae containing three discrete clades ( Pseudobiceros + Thysanozoon ; Phrikoceros + Tytthosoceros ; Pseudoceros ) is recovered, and in each clade we are able to resolve existing taxonomic uncertainties and junior synonyms (see the three subsections on Pseudoceros splendidus , Phrikoceros and Tytthosoceros and Pseudoceros below). Species characterized by two male reproductive systems ( Pseudobiceros and Thysanozoon ) cluster together, with each genus forming a monophyletic group. This is in contrast to the findings of Bahia et al. (2017), who recovered paraphyletic Pseudobiceros and Thysanozoon . Based on a grouping of their specimen of Pseudobiceros hancockanus ( Newman & Cannon, 1994) in Pseudoceros, Tsunashima et al. (2017) also questioned the monophyly of Pseudoceros and Pseudobiceros . However, this might be attributable to a misidentification ( Tsunashima et al. 2017: fig. 2H on p. 163 is not Pseudobiceros hancockanus ). Our specimen of Pseudobiceros hancockanus clusters with Pseudobiceros . Our analysis also includes two specimens of an undescribed species of Pseudobiceros . The colour pattern of this species (black dorsal surface, with inner wide, bright orange and outer narrower, white marginal bands encircling the entire body) led Newman & Cannon (1994: fig. 51C; 2003: 33, 39, 52, 82) to identify it as Pseudobiceros hancockanus . Accordingly, this description is widely circulated in online databases. However, the original description of Pseudobiceros hancockanus by Collingwood (1876) differs considerably. Recently, Bolaños et al. (2016) have addressed the confusion surrounding Pseudobiceros hancockanus and have designated a neotype for the species.

Our results support the duplication of the male reproductive system as a valid morphological trait ( Rawlinson & Litvaitis, 2008), separating the genera Pseudobiceros and Thysanozoon from the remaining pseudocerotids. However, species resolution in Thysanozoon was not possible using the D1–D2 expansion segments. In fact, morphologically distinct species [e.g. Thysanozoon brocchii (Risso, 1818) and Thysanozoon raphaeli Bolaños et al., 2007 ] reveal almost complete sequence concordance. Consequently, specific separations in this genus await analyses that use more rapidly evolving genetic markers (possibly COI, internal transcribed spacers).

Pseudobiceros splendidus ( Lang, 1884) View in CoL ( Fig. 9; Table 2): Pseudobiceros splendidus View in CoL , initially described as Pseudoceros superbus View in CoL by Lang (1884), was renamed Pseudobiceros splendidus View in CoL by Stummer-Traunfels (1933) to avoid confusion with another species with the same name. Eventually, it was moved to Pseudobiceros View in CoL by Faubel (1984), who recognized its double male copulatory system. Pseudobiceros splendidus View in CoL is part of a group of pseudocerotids exhibiting colour patterns with a uniform dorsal colour plus one or more marginal bands of a different colour ( Newman & Cannon, 1994, 1997). In most Pseudobiceros splendidus View in CoL , the dorsal colour is black, and the animals possess an orange submarginal band and a black rim ( Fig. 9A; Table 2). Lang’s (1884: 541) original description mentioned tiny white dots on the dorsal surface that can be discerned only with a magnifying glass, but Hyman (1955a) and Prudhoe (1989) made no mention of these minute white dots.

Pseudobiceros hymanae Newman & Cannon, 1997 View in CoL also has a dorsal black coloration, a submarginal orange band and a black margin ( Newman & Cannon, 1997). The authors justified a new species designation because the orange band tends towards a rusty colour ( Fig. 9B; Table 2), and they placed greater importance on the microscopic white dots found on Pseudobiceros splendidus View in CoL ( Table 2) than on the marginal band. Depending on the nutritional state or the lifehistory stage of a specimen, Pseudobiceros hymanae View in CoL with a reddish-brown dorsal colour have also been described ( Bolaños et al., 2016: 150). Furthermore, reddish-brown specimens with a submarginal orange band surrounded by a black rim are reminiscent of Pseudobiceros evelinae ( Marcus, 1950) View in CoL ( Table 2; Bahia et al., 2012: 38; Bahia et al., 2014: 507), yet another colour morph similar to Pseudobiceros splendidus View in CoL .

Newman & Cannon (1994) described two additional, similarly coloured species. They defined Pseudobiceros periculosus Newman & Cannon, 1994 View in CoL as distinct from Pseudobiceros splendidus View in CoL because of the lack of a black rim ( Fig. 9C; Table 2). Instead, an extremely thin transparent margin surrounds the animal. However, this might be attributable to the nutritional status of the specimen or might represent an artefact of light refraction when taking photographic records through the air–water interface. The second species, Pseudoceros periaurantius Newman & Cannon, 1994 View in CoL , has a wide orange marginal band extending all the way to the rim ( Fig. 9D), and because of its single male copulatory complex, clearly is placed in Pseudoceros View in CoL not Pseudobiceros View in CoL . Mitochondrial sequences showing Pseudobiceros periaurantius clustering in Pseudoceros View in CoL add further support to its taxonomic placement ( Aguado et al., 2017).

A comparison of D1–D2 expansion segment sequences of the two specimens of Pseudobiceros splendidus we collected in Florida with Pseudobiceros periculosus from the Great Barrier Reef reveals that they are identical. Furthermore, Pseudobiceros evelinae sequences available in GenBank are identical to our Pseudobiceros periculosus and Pseudobiceros splendidus sequences. Therefore, based on colour patterns and nucleotide sequences of the 28S D1– D2 expansion segments, these species can no longer be maintained as distinct entities. We conclude that Pseudobiceros periculosus , Pseudobiceros hymanae and Pseudobiceros evelinae are all junior synonyms of Pseudobiceros splendidus . A review of the geographical distribution of Pseudobiceros splendidus reveals that it might be one of the few truly cosmopolitan polyclad species ( Table 2) and a testament to strong intraspecific cohesion.

Phrikoceros Newman & Cannon, 1996 View in CoL and Tytthosoceros Newman & Cannon, 1996 View in CoL : The second pseudocerotid clade forms an immediate sister group to Pseudobiceros View in CoL + Thysanozoon View in CoL and includes Phrikoceros View in CoL and Tytthosoceros View in CoL . Like species of Pseudoceros View in CoL , both genera possess single male reproductive systems. However, their overall gross morphologies (e.g. highly ruffled body margin, shape of pseudotentacles, ruffled pharynx with simple folds) resemble Pseudobiceros ( Newman & Cannon, 1996a) View in CoL . These differences led Newman & Cannon (1996a) to establish the new genus Phrikoceros View in CoL . A further separation places species with ear-like pseudotentacles into Tytthosoceros View in CoL , while species with square pseudotentacles are retained in Phrikoceros ( Newman & Cannon, 1996b) View in CoL . Based on our results, this separation is artificial. Given that the establishment of Phrikoceros ( Newman & Cannon, 1996a) View in CoL pre-dated Tytthosoceros ( Newman & Cannon, 1996b) View in CoL , we here abolish Tytthosoceros View in CoL as a genus and move all three valid species to Phrikoceros View in CoL to establish Phrikoceros nocturnus View in CoL comb. nov., Phrikoceros lizardensis View in CoL comb. nov. and Phrikoceros inca View in CoL comb. nov.

Given that Bahia et al. (2017) found that Phrikoceros View in CoL clustered with other genera characterized by double male reproductive systems, they invoked the possibility that the genus has two male systems opening into a single male gonopore. However, not only does the original description of Phrikoceros View in CoL show a single male system ( Newman & Cannon 1996a), but our own histological examinations of Phrikoceros mopsus ( Marcus, 1952) View in CoL also reveal only one male system.

P s e u d o c e r o s L a n g, 1 8 8 4: T h e t h i r d c l a d e i n Pseudocerotidae is formed by Pseudoceros . Although the genus is monophyletic, we find substructure in the group. With only a few exceptions, most of our samples were collected from Australia, especially the Great Barrier Reef, where they represent a conspicuous component of the reef fauna ( Newman & Cannon, 2003 and references therein).

The type locality for Pseudoceros prudhoei Newman & Cannon, 1994 is Heron Island, Great Barrier Reef, Australia ( Newman & Cannon, 1994). The species has been recorded throughout the Indo-Pacific, including Japan, the Persian Gulf and Kenya. The original species description notes a brown-orange dorsal background colour with a wide, inner light mauve to light purple marginal band and a narrow, outer yellow band ( Newman & Cannon, 1994). Recently, Velasquez et al. (2018) recorded a much darker colour variant (even black dorsal colour, inner marginal band almost white) from the Israeli coast of the Mediterranean. Based on comparisons with the original description of Pseudoceros duplicinctus Prudhoe, 1989 , the authors synonymized Pseudoceros prudhoei with Pseudoceros duplicinctus . By comparing the sequence of a specimen identified as Pseudoceros prudhoei with the existing sequence of Pseudoceros duplicinctus , we here lend molecular support to this synonymy. Additionally, Velasquez et al. (2018) entertained the possibility that Pseudoceros depiliktabub Newman & Cannon, 1994 is also a junior synonym of Pseudoceros duplicinctus . Our comparison of sequences for a specimen initially identified as Pseudoceros depiliktabub with that of Pseudoceros duplicinctus confirmed this hypothesis, despite different colour variations and different collection localities of the specimens. We here designate Pseudoceros depiliktabub a junior synonym of Pseudoceros duplicinctus .

Only a handful of Pseudoceros species have been reported from the Caribbean ( Pseudoceros pardalis , Pseudoceros aurolineatus Verrill, 1901 ; Pseudoceros splendidus , Pseudoceros bicolor and Pseudoceros rawlinsonae ). In our survey, we found Pseudoceros splendidus (see above) and a species complex of Pseudoceros bicolor exhibiting several colour morphs ( Litvaitis et al., 2010). We also collected several specimens of Pseudoceros pardalis , a species that is characterized by a distinct colour pattern. After careful examination of its reproductive system, we had reassigned the species to Pseudobiceros ( Bolaños et al. 2007) . This reassignment now has been confirmed based on 28S rDNA sequences.

TAXONOMIC VALUE OF MORPHOLOGICAL TRAITS AND SUMMARY OF TAXONOMIC REVISIONS

Our results confirm that, with few exceptions, the cotyl (sensu Lang, 1884) is a taxonomically useful character in the subordinal division of Polycladida . We encourage its use for the placement of species into Cotylea vs. Acotylea. Other traits that are valid for separating most species into Acotylea and Cotylea include the orientation of the male reproductive system and the presence of Lang’s vesicle ( Faubel , 1983, 1984). Exceptions to the taxonomic usefulness of these characters are found in basal cotylean lineages (e.g. Cestoplanidae , Boniniidae and Theamatidae ). Based on these characters, the basic acotylean body plan then lacks a cotyl, has a male system located anterior to the male gonopore and possesses Lang’s vesicle. Additionally, all Acotylea have a ruffled pharynx.

In earlier, morphology-based classification systems, the most reliable taxonomic characters used had been associated with the reproductive systems ( Faubel , 1983, 1984), the arrangement of eyes and the position and shape of tentacles ( Prudhoe, 1985). Despite Faubel’s (1983, 1984) attempt to distinguish acotylean superfamilies based on one of three types of prostatic vesicles, such separation is no longer supported ( Table 1) and requires the inclusion of additional characters. Faubel (1983) united Stylochoidea by the ‘presence of a free prostatic vesicle’. However, because Hoploplanidae with their atypical, interpolated prostatic vesicles are included in the superfamily, a free prostatic vesicle can no longer be considered a synapomorphy for the superfamily. Alternatively, four-lobed Götte’s larvae are taxonomically widespread in Stylochoidea (inclusive of Hoploplanidae ). However, rather than use their presence as a stylochoid synapomorphy, we propose that the ancestral polyclad had a biphasic life cycle. Stylochoidea (sensu Poche, 1926) lacks a morphology-based synapomorphy at this time and is supported only by molecular data.

In Cryptoceloidea (sensu Bahia et al., 2017), Ilyplanidae View in CoL , Discocelidae View in CoL and Euplanidae View in CoL lack a true prostatic vesicle. However, Cryptocelidae View in CoL possess an interpolated prostatic vesicle. Consequently, using the type of prostatic vesicle alone is taxonomically not informative. Three of the five recognized cryptocelid genera are monospecific ( Faubel , 1983) and poorly defined. The two species-rich genera Cryptocelis View in CoL and Phaenocelis View in CoL possess unarmed, rod- or cylindrical-shaped penis papillae. We contend that Cryptoceloidea is united by the ‘absence of a true prostatic vesicle or, if an interpolated prostatic vesicle is present, an unarmed conical penis papilla’.

Finally, a penis armed with cuticular elements (e.g. stylet, cirrus) is a synapomorphy of Leptoplanoidea View in CoL (sensu Faubel , 1983, 1984), and the absence of armature in Notocomplanidae View in CoL represents most probably a secondary loss ( Table 1). Previous studies have also used the shape/amount of folding of the epithelial lining of the prostatic vesicle and the degree to which the ejaculatory duct extends into the vesicle as diagnostic characters ( Faubel , 1983; Bulnes et al., 2005). We argue that such characters can be subjective and are mostly dependent on the quality of fixation and the angle of the plane of sectioning. Hence, we advise caution as to their taxonomic use.

The shape of marginal tentacles provides taxonomic information in Cotylea ( Faubel , 1984; Prudhoe, 1985; Rawlinson & Litvaitis, 2008). Pseudotentacles are formed by a ruffling of the anterior margin and are a characteristic of all Pseudocerotidae . Their shape may range from simple folds to more intricate structures ( Newman & Cannon, 1994). The presence or absence of pseudotentacles is a useful trait, whereas the degree and complexity of folding (e.g. simple folds, ear-like or square) are dependent on the size of the animal, may be species specific and are affected by the quality of fixation. Consequently, only their presence should be noted, establishing an apomorphy for Pseudocerotidae ( Table 1). Cotylean tentacles can also extend as well-separated, pointed structures from the anterior margin ( Diposthus and Pericelis ), arise as fine lateral tentacles ( Boniniidae ), form small marginal bumps ( Cycloporus ) or form pointed, V-shaped extensions ( Maritigrella and Eurylepta ). Thus, marginal tentacles can be useful for generic separation in Cotylea.

Given that the position and arrangement of marginal eyes are highly variable in polyclads and can vary over the lifetime of an individual ( Faubel , 1983, 1984; Prudhoe, 1985), the trait, at best, should be considered only in mature adult animals. The arrangement of pseudotentacular eyes has been used for generic separation in Pseudoceros vs. Pseudobiceros ( Newman & Cannon, 1994, 1996a). However, they are often difficult to discern in animals of dark background colour. The number of male gonopores (i.e. Pseudoceros , one; Pseudobiceros , two) and the extent of folding of the pharynx are more reliable characters. Pharynx shape (ruffled vs. cylindrical) led Faubel (1984) to establish Pseudocerotoidea and Euryleptoidea, two superfamilies that can no longer be supported with the present data available ( Rawlinson & Litvaitis, 2008; this study). Cotylean pharynx shape may be used to identify families in conjunction with additional traits.

Our results supported many portions of earlier classifications based on morphology and molecular data, but we also identified several taxonomic corrections and additions. The following is a summary of those changes. In Acotylea, we returned Phaenoplana peleca to its original genus Phaenocelis ( Phaenoplana peleca is now a junior synonym of Phaenocelis peleca ), we established the new family Notocomplanidae to unite notoplanids lacking penis armature and, at the same time, transferred Melloplana ferruginea to Notocomplana ferruginea , resulting in the elimination of the genus Melloplana . Furthermore, we also returned Pleioplana atomata to its original Notoplana atomata ( Pleioplana atomata is now a junior synonym of Notoplana atomata ) and transferred Persica qeshmensis to Notoplana qeshmensis , resulting in the abolishment of the monospecific genus Persica . Finally, our results indicated that Amyris ujara is a junior synonym of Amyris hummelincki .

Changes in Cotylea also included rearrangements of taxa and the identification of junior synonyms. We designated Chromoplanoidea (sensu Bahia et al., 2017) as invalid because the type family is not contained in the superfamily.As a consequence, Boniniidae , Theamatidae and Amyellidae remain their own distinct families. We amended Diposthidae to include all species of Pericelis , resulting in the elimination of the family Pericelidae and the invalidation of Periceloidea (sensu Bahia et al., 2017). Furthermore, we identified the following synonyms and new combinations: Boninia divae is a junior synonym of Boninia antillarum ; Cestoplana australis is a junior synonym of Cestoplana rubrocincta ; Maritigrella crozieri has been moved to Prostheceraeus crozieri ; Maritigrella newmanae is a junior synonym of Prostheceraeus floridanus ; Amakusaplana acroporae has been moved to Prosthiostomum acroporae ; and Lurymare utarum has been returned to its original genus, Prosthiostomum , as Prosthiostomum utarum . In addition, the genus Tytthosoceros has been abolished, and all three valid species have been moved to Phrikoceros to establish Phrikoceros nocturnus , Phrikoceros lizardensis and Phrikoceros inca . Finally, Pseudoceros prudhoei and Pseudoceros depliktabub have been identified as junior synonyms of Pseudoceros duplicinctus , and Pseudobiceros evelinae , Pseudobiceros periculosus and Pseudobiceros hymanae are all junior synonyms of Pseudobiceros splendidus .

With the application of innovative methodological approaches (e.g. DNA sequencing, phylogenomics, fluorescence microscopy and immunocytochemistry), a more stable system for Polycladida is emerging. Although multiple lines of evidence are supporting three acotylean superfamilies and a cotylean classification based on families, the relationships in some clades require additional research. Specifically, the families Notoplanidae and Stylochoplanidae (both in Leptoplanoidea ) need revision. Likewise, relationships in Euryleptidae , and the monophyly of this family itself, need further resolution. Finally, the relationships of basal cotylean lineages (i.e. Boniniidae , Theamatidae and Amyellidae ) deserve additional attention.