Unravelling the convoluted nomenclature of Marphysa simplex (Annelida, Eunicidae) with the proposal of a new name and the re-description of species

Marphysa simplex is a name that three species bear within the same genus, but each has a different authority and morphological characteristics. This homonymy condition leads to taxonomic confusion and the finite designation of name-bearing is imperative. The current study focuses on two species identified as M. simplex Crossland, 1903 and M. simplex Treadwell, 1922 and a third one, recently considered a secondary homonymy, M. simplex (Langerhans, 1884), is also assessed. The available type specimens were examined and re-described in detail using updated characters and the original descriptions. Marphysa simplex (Langerhans, 1884) is herein judged as an indeterminable species. Marphysa simplex Crossland, 1903 is confirmed as a junior synonym of M. teretiuscula (Schmarda, 1861a) because the differences are minimal. Moreover, M. teretiuscula has characteristics similar to Group B2 (Sanguinea-group; only compound spinigers), instead of the Teretiuscula-group (compound spinigers in the anterior region, subacicular limbate in all chaetigers). On the other hand, M. simplex Treadwell, 1922 is a junior primary homonym of Crossland’s species replaced by M. fijiensis nom. nov. with the chaetal arrangement similar to Group A (limbate chaetae only). In conclusion, the name M. simplex is now unacceptable. The hypothesis on species group only with limbate chaetae and the redescription on M. teretiuscula is also given.


Introduction
Within the long story of Marphysa de Quatrefages, 1865, the name M. simplex has been referred to in several homonymy cases since three species worldwide bear this name: Marphysa simplex Crossland, 1903, M. simplex Treadwell, 1922and M. simplex (Langerhans, 1884. Simultaneously, the names M. simplex Crossland, 1903 andTreadwell 1922 have been used as synonymies of two other Marphysa species (Glasby and Hutchings 2010). In this study, we provide an analysis of these taxonomic issues.
The name M. simplex was firstly used by Crossland (1903) for two specimens from Zanzibar (Tanzania). Although Crossland proposed a new name, he also stated that his specimens resemble M. teretiuscula (Schmarda, 1861a) from Sri Lanka because both species are present in the Indian Ocean. Nonetheless, he could not make an accurate comparison because Schmarda's description was short and poorly illustrated (Crossland 1903;Glasby and Hutchings 2010).
About two decades later, Treadwell (1922) described a new species, M. simplex, based on one specimen found together with M. macintoshi Crossland, 1903 in Suva Harbour, Fiji. Even though Treadwell knew about Crossland's contributions, it is uncertain why he used an already established name for his new species.
Later, a third species with the same name appeared as a secondary homonym (ICZN 1999, Art. 57.3). Langerhans (1884) described Amphiro simplex, based on a single specimen from Madeira (Portugal). Amphiro Kinberg, 1865 was considered a junior synonym of Marphysa by Hartman (1949); therefore, the species described in that genus became part of Marphysa, including Langerhans' species.
The three species named M. simplex differ in morphology and can be classified into three of the informal groups proposed by Fauchald (1970). Crossland's specimens belong to Marphysa B2 group (with compound spinigers only), Treadwell's specimens belong to A group (without compound chaetae) and Langerhans's species to group C (with compound falcigers only).
Currently, the species M. simplex (Langerhans) is considered indeterminable by Molina-Acevedo and Idris (accepted, but not yet published) since the type material is lost and the original description lacks enough diagnostic characters for comparison, even if new topotypes are found. The other two species of M. simplex were accepted as synonyms of different Marphysa species. Glasby and Hutchings (2010), who commented about these homonymies, considered M. simplex Crossland a junior synonym of M. teretiuscula following Crosland's comments regarding the similarity and also because the body shape and the length of the peristomial appendages were consistent between both species. Likewise, Glasby and Hutchings (2010) synonymised M. simplex Treadwell with M. mossambica (Peters, 1854) because the only differences between both species were size-related variations.
In the present study, we confirm the synonymy of M. simplex Crossland and M. teretiuscula after a detailed evaluation of the type material. However, we propose the re-establishment of M. simplex Treadwell, as it is distinguished morphologically from M. mossambica. Since M. simplex Treadwell is a primary homonym of Crossland's species (ICZN 1999, Art. 53.3) and therefore no longer valid, we hereby replace Treadwell's name by erecting Marphysa fijiensis nom. nov. (ICZN 1999, Art. 60.3). Likewise, we provide re-descriptions of M. teretiuscula and M. fijiensis nom. nov., based on the type materials, some observations regarding group A (Fauchald 1970) and a group of Marphysa species with compound spinigers and subacicular limbate chaetae (Glasby and Hutchings 2010). Comparative Tables with species similar to M. fijiensis nom. nov. and M. teretiuscula are included.

Material and methods
The materials reviewed in this study were deposited in the following institutions: American Museum of Natural History, New York, USA (AMNH), Australian Museum (AM), The Natural History Museum, London (BNHM), Natural History Museum, Vienna, Austria (NHMW), National Museum of Natural History, Smithsonian Institution, Washington D.C., USA (USNM) and Museum of Natural History, Berlin, Germany (ZMB).
The re-descriptions of species were illustrated with digital photographs. A series of photos were stacked using Hel-iconFocus 6 (Method A) software to improve the depth of field and the final images were edited and assembled into plates using Adobe Photoshop 2020. The re-descriptions include: prostomium, body, branchiae, maxillary apparatus, parapodia, compound chaetae and simple chaetae. The terminology to describe the overall morphology of characters followed those recently provided by Carrera-Parra (2015, 2017), Zanol et al. (2016) and Molina-Acevedo (2018). Paired and unpaired maxillae were indicated as 'M' followed by a Roman number (e.g. MI, MII). The maxillary apparatus and the pectinate chaetae were described according to Carrera-Parra (2015, 2017) and Zanol et al. (2016), respectively. In addition, the definition of long and short branchial filaments, proposed by Molina-Acevedo and Idris (2020), is followed. Three or five parapodia per species were dissected to compare the different shapes of the parapodial cirri, lobes and chaetae throughout the body.
Some specimens studied were posteriorly incomplete; hence, morphological measurements for the length up to chaetiger 10 (L10) and width at chaetiger 10 (W10), excluding parapodia, were used as a size estimate of an individual worm. Additionally, the total length (TL) and the total number of chaetigers (TChae), the chaetiger number and side (R for right, L for left) from which branchiae and subacicular hooks emerged, were also recorded. Furthermore, for specimens from M. teretiuscula, linear regression analyses were included to evaluate the relationships between L10, W10 and several morphological features, including the starting of branchiae and subacicular chaetae and the last chaetiger with developed postchaetal lobe. The degree of predictability of variation (coefficient of determination) in the features according to the sizes is given as R 2 (e.g. R 2 = 0.7, p = 0.05, n = 8). Marphysa simplex Treadwell, 1922: 151-152, pl. 5, figs 8-12, text-figs 39 (non Crossland, 1903; -Hartman 1956: 254, 262, 268, 286. Material examined. Holotype. Fiji · Suva Harbour; Apr-Jun, 1920; AMNH 1920-1530.
Peristomium (1.7 mm long, 2.3 mm wide) larger than prostomium, first ring twice as long as second ring, separation between rings distinct on all sides ( Fig. 1A-C). Ventral lip dissected, with several shallow wrinkles (Fig. 1C).
Branchiae pectinate with up to five long filaments, from chaetigers 22 to 2D,E). Six first branchiae with one filament; reaching maximum five filaments in chaetigers 79L-178L; last 12 branchiae with one filament (Fig. 3A). Branchial filaments longer than dorsal cirri, except in first six and last 10 branchiae.
Distribution. Known only from the type locality.
Habitat. Uncertain. Possibly coral reefs or mudflats (Treadwell 1922). Remarks. The first comment on the synonymy of Marphysa fijiensis nom. nov. (as M. simplex Treadwell, 1922) and M. mossambica was made by Hartman (1956). Subsequently, Glasby and Hutchings (2010) supported this idea, stating that the morphology described by Treadwell was very similar to the smaller-sized specimens of M. mossambica. However, after a detailed morphological comparison of both species' type material, we found marked differences. Marphysa fijiensis nom. nov. lacks eyes, whereas they are present in M. mossambica. Additionally, M. fijiensis nom. nov. (L10: 8.2 mm) has a maximum number of five branchial filaments in the median region (Fig. 3), whereas M. mossambica (L10:8.5-11.5 mm), has a maximum number of eight branchial filaments in the mid-posterior region (Fig. 3). In M. fijiensis nom. nov., the chaetal lobe is rounded in the anterior region and the postchaetal lobe is oval in the first three chaetigers. In contrast, in M. mossambica, the chaetal lobe is rectangular in the anterior region and the postchaetal lobe is digitiform in the first three chaetigers.  (Kinberg, 1865) in lacking compound chaetae. However, M. fijiensis nom. nov. lacks the peduncle in prostomial appendages, with swollen base in ventral cirri starting from chaetiger 14 and the acicula is twice as wide as the subacicular hook in the median-posterior region. In contrast, M. moribidii (L10: 12.2-20 mm) has a peduncle at the base of the palpo-and ceratostyles, bears ventral cirri  Fauchald (1970). Abbreviations: MF: Maxillary formula, roman numerals refer to number of maxilla; MxC: maxillary carriers; CIS: closing system; COp: cavity opening; PR-I: first peristomial ring; PR-II: second peristomial ring; Chaet: chaetiger; p/a: present/absent; AR: anterior region; MR: median region; PR: posterior region; SH: subacicular hook. INSS: Isodont narrow with short and slender teeth; INLS: Isodont narrow with long and slender teeth; IWSS: Isodont wide with short and slender teeth; IWLS: Isodont wide with long and slender teeth; AWLT: Anodont wide with long and thick teeth. with a swollen base starting from chaetiger 6 and has a subacicular hook similar in width to the acicula throughout the body. Furthermore, M. fijiensis nom. nov. has the prechaetal lobe as a transverse fold throughout the body, the chaetal lobe rounded throughout the body, the ventral cirri with a swollen base starting from chaetiger 14 and the subacicular hook starting from chaetiger 25; while M. novaehollandiae (L10: 6.6-9.6 mm) has the prechaetal lobe with dorsal edge longer than the ventral side in the first chaetigers, the chaetal lobe rectangular in the anterior region, the ventral cirri with a swollen base starting from chaetiger 8 and the subacicular hook starting from chaetigers 39-42. The comparison of M. fijiensis nom. nov. with related species is provided in Table 1.  Fig. 5A, B, D) shallow and deep ventrally. Prostomial appendages in semicircle, median antenna isolated by a gap. Palps reaching second chaetiger; lateral antennae reaching middle of third chaetiger; median antennae reaching fourth chaetiger. Palpophores and ceratophores ring-shaped, short, thick; palpostyles and ceratostyles tapering, slender. On right side with two palpostyle in the same palpophore (Fig. 5C). Eyes oval, brown, between palps and lateral antennae.
First pair of parapodia small; best developed in chaetigers 11-56, following parapodia gradually decreasing in size. Dorsal cirri conical in all chaetigers; longer than ventral cirri in anterior and posterior chaetigers, shorter in median chaetigers; best developed in chaetigers 3-37, following gradually decreasing in size (Fig. 6A-K). Prechaetal lobes short, in anterior chaetigers dorsal edge longer than ventral, in median-posterior chaetigers, as transverse  fold (Fig. 6A-K). Chaetal lobes rounded in first 35 chaetigers, shorter than other lobes, with aciculae emerging dorsal to mid-line; triangular from chaetiger 36, longer than other lobes, with aciculae emerging in mid-line (Fig. 6A-K). Postchaetal lobes well developed in first 56 chaetigers; ovoid with dorsal edge longer than ventral edge in following chaetigers; progressively smaller from chaetiger 19; from chaetiger 57, inconspicuous (Fig. 6A-K). Ventral cirri conical in first five chaetigers; from chaetiger 6 to last chaetiger of fragment with short oval swollen base and digitiform tip (Fig. 6A-K). Second syntype with ventral cirri with short oval swollen base and digitiform tip up to 27 chaetigers before pygidium; digitiform in following ones, gradually decreasing in size posteriorly.
In second syntype, pygidium with dorsal pairs of anal cirri, as long as last 12 chaetigers; ventral pair of anal cirri short, as long as last three chaetigers (Fig. 5G).
Habitat. Unknown. Schmarda (1861a) did not indicate the habitat of the species.

Discussion
Marphysa group without compound chaetae At present, the small Marphysa group A proposed by Fauchald (1970) consists of four species characterised by the absence of compound chaetae: Marphysa mossambica from Mozambique, M. novaehollandiae from Sydney, M. moribidii from Malaysia and M. fijiensis nom. nov. from Fiji. All these species are represented by large organisms with a high number of segments (more than 200) in the adult stage. However, the ontogenetic development of these species and the presence of compound chaetae in the early stages, a common condition in other Marphysa species (Southern 1921;Aiyar 1931;Pillai 1958), are unknown.
The absence of compound chaetae was an important character to consider the species in the Marphysa group A as an independent genus, Nauphanta Kinberg, 1865(Fauchald 1987. However, Glasby and Hutchings (2010) regarded Nauphanta as a junior synonym of Marphysa, based on the variation in that feature in ontogeny. In addition, Zanol et al. (2014) supported this synonymy in a phylogenetic analysis of Eunicidae, based on morphological and molecular evidence, confirming this uncommon condition is developed in some Marphysa species.
The emergence of compound chaetae in the early stages has been well documented in some Marphysa species from India: Marphysa borradailei (Borradaile 1902;Pillai 1958), M. gravelyi (Malathi et al. 2011) and Marphysa sp. (Aiyar 1931). During the development of the first chaetiger in the metatrochophore larvae of these species, the first compound chaeta is characterised by having convex and straight edges as they emerge (Borradaile 1902;Southern 1921;Aiyar 1931;Pillai 1958) just after the appearance of two unjointed (simple) chaetae (Borradaile 1902;Aiyar 1931;Malathi et al. 2011). This compound chaeta resembles the blade of the compound falciger, although it lacks the typical distal teeth. When the third chaetiger is developed, only the same compound chaeta emerges in chaetigers 2 and 3 (Borradaile 1902;Southern 1921;Aiyar 1931;Pillai 1958). At the time of development of chaetigers 12-13, two types of compound chaetae can be observed (Aiyar 1931;Pillai 1958): (1) a hooked chaeta, herein judged as the compound falciger; and (2) a sickle-shaped chaeta, also called falcigerous by Aiyar (1931) and Pillai (1958), herein interpreted as the typical compound spiniger. Aiyar (1931) also described the shift and loss of chaetae in larger specimens of Marphysa sp. The specimens have limbate chaetae throughout the body regardless of size, in Table 2. Morphological features of Marphysa species with reddish subacicular hook and three types of pectinate chaetae. Abbreviations: MF: Maxillary formula, roman numerals refer to number of maxilla; MxC: maxillary carriers; CIS: closing system; COp: cavity opening; PR-I: first peristomial ring; PR-II: second peristomial ring; Chaet: chaetiger; p/a: present/absent; AR: anterior region; MR: median region; PR: posterior region; SH: subacicular hook. INSS: Isodont narrow with short and slender teeth; INLS: Isodont narrow with long and slender teeth; INLT: Isodont narrow with long and thick teeth; IWSS: Isodont wide with short and slender teeth; IWLS: Isodont wide with long and slender teeth; IWLT: Isodont wide with long and thick teeth; AWLT: Anodont wide with long and thick teeth; AWLS: Anodont wide with long and slender teeth.   contrast to the compound chaetae lost to different extents.

M. americana
The compound spinigers are present in all chaetigers in specimens with up to 150 chaetigers, but they disappear in the most-posterior chaetigers in larger specimens; on the contrary, the compound falcigers appear only in the posterior region of specimens with 70-80 chaetigers and entirely disappear in specimens with 150 or more chaetigers. Glasby and Hutchings (2010) indicated that the lack of compound falcigers in Marphysa is due to their loss during the transition from juvenile to young adults, as occurs in M. angelensis Fauchald, 1970 (Molina-Acevedo and Villalobos-Guerrero, in prep.), M. borradailei (Southern 1921;Pillai 1958), M. brevitentaculata Treadwell, 1921 (Salazar-Vallejo andCarrera-Parra 1998; Molina-Acevedo and Carrera-Parra 2017), M. gravelyi (Southern 1921;Pillai 1958) and M. sherlockae Kara et al. 2020(Kara et al. 2020. Likewise, Glasby and Hutchings (2010) stated that the absence of compound spinigers is a paedomorphic condition in Marphysa group A since they are supposedly lacking in the juveniles. However, it contradicts the evidence demonstrated by Aiyar (1931) and Pillai (1958) for juveniles of Marphysa sp. and M. borradailei having compound spinigers.
The absence of both compound spinigers and falcigers in species of Marphysa group A is more likely due to their gradual loss as the body of the specimens increases in size. The small juveniles have both compound spinigers and falcigers, but the latter chaetae begin to disappear as the animal grows. In juveniles or young adults, the falcigers may be lost entirely, with the compound spinigers being replaced by limbate chaetae. Finally, in adults, both compound spinigers and falcigers may be lacking. Several suitable examples allude to this chaetal transition. For instance, in the group with limbate subacicular chaetae sensu Glasby and Hutchings (2010, see below), the adults have compound spinigers in the anterior-median region, which are replaced by limbate chaetae in the median-posterior. Likewise, in M. brevitentaculata, M. digitibranchia Hoagland, 1920 andM. mangeri Augener, 1918, the compound falcigers are present solely in the anterior region, the compound spinigers only in the anterior-median region and limbate subacicular chaetae only in median-posterior region (Augener 1918;Hoagland 1920; Molina-Acevedo and Carrera-Parra 2017). To confirm this hypothesis, a thorough ontogenetic study on the chaetal shift is needed to understand the absence of compound chaetae in Group A.

Marphysa teretiuscula, a misinterpreted species
Marphysa teretiuscula has received little attention since Schmarda (1861a) described it initially from the east coast of Sri Lanka (formerly Ceylon). Although de Quatrefages (1866), Ehlers (1868) and Crossland (1903) recognised the species as valid in Marphysa, it was overlooked by some authors who extensively studied the polychaetous annelids from India, such as Gravely (1927) and Fauvel   (1930,1932,1940,1953). Marphysa teretiuscula was briefly mentioned in the early 1900s when Borradaile (1902) described the larval development of specimens from the north of Sri Lanka that release the embryos in jelly cocoons attached to the bottom by a stem. One of these worms was identified by Arthur Willey as allied to, or identical with, M. teretiuscula (see Borradaile 1902), the only species known at that time from the area. Later, Southern (1921) and Aiyar (1931) found a specimen with similar reproductive patterns to those described by Borradaile, which were identified as M. gravelyi (Southern) from the Chilka Lake and Marphysa sp. from Madras. Pillai (1958) studied the specimens from the Negombo Lagoon, Sri Lanka, which also fitted in terms of ontogeny to M. teretiuscula sensu Borradaile (1902), but differed from Schmarda's original description in several morphological respects. In this regard, Pillai (1958) described these specimens as a new species, M. borradailei. Regrettably, the reproduction and development of M. teretiuscula are still unknown.
Marphysa teretiuscula was chosen by Glasby and Hutchings (2010) to create the informal 'Teretiuscula-group' within Marphysa, which is featured by having species with compound spinigers restricted to anterior chaetigers and subacicular limbate chaetae present throughout the body, in addition to the formation of jelly cocoons as part of a reproductive strategy. The species included in the 'Teretiuscula-group' are M. fauchaldi, M. borradailei, M. gravelyi, M. teretiuscula (Glasby and Hutchings 2010) and the recently described species, M. madrasi (Hutchings et al. 2020). However, the stem species M. teretiuscula, in fact, lacks those relevant features encompassing the 'Teretiuscula-group'. As demonstrated in the present study's re-description, M. teretiuscula bears compound spinigers only and lacks subacicular limbate chaetae in all parapodia (Fig. 7E, F). Furthermore, the reproduction of M. teretiuscula is unknown compared with those species from the same region, M. borradailei and M. gravelyi. Herein, we judge M. teretiuscula as belonging to the group B2 sensu Fauchald (1970), whereas the 'Teretiuscula-group' sensu Glasby and Hutchings (2010) is renamed with the earliest species described within, 'Gravelyi-group'.
Two other species have also been described from near the type locality of M. teretiuscula: Paramarphysa orientalis Willey, 1905 andM. chevalensis Willey, 1905, both from the Gulf of Manaar (India). The first species is recently considered indeterminable by Molina-Acevedo and Idris (accepted, but not yet published), while the second species is here treated in the same manner. Marphysa chevalensis was described, based on two specimens: the smaller has both compound falcigers and spinigers in all chaetigers, whereas the larger has a similar chaetal pattern; however, the falcigers are restricted to the anterior-median region (Willey 1905). The larger specimen could have undergone a loss of falcigers as demonstrated in the young adults of M. gravelyi, M. brevitentaculata and M. sherlockae (Southern 1921;Pillai 1958;Sala-zar-Vallejo and Carrera-Parra 1998;Kara et al. 2020). Marphysa chevalensis may be the young adult of one of the species described from Sri Lanka or India (M. gravelyi, M. borradailei and M. teretiuscula) that has not yet developed its adult complement of chaetae. On the other hand, the type material of M. chevalensis is lost and relevant diagnostic characters, such as the shape of parapodia, the maxillary formula, the presence of subacicular limbate chaetae, the colour of the subacicular hook, among others, were not included in the original description. This lack of information hinders the differentiation between similar species, even when fresh topotype material is available. Therefore, we also consider M. chevalensis as an indeterminable species.

Conclusion
In the present work, we clarified the taxonomic status of five Marphysa species. The synonymy of M. teretiuscula over M. simplex Crossland was confirmed and the re-description of the species was provided using the type material. Marphysa simplex Treadwell was re-established, re-described and a new name for this homonym species was proposed, M. fijiensis nom. nov. Likewise, M. simplex (Langerhans), M. chevalensis and Paramarphysa orientalis were considered indeterminable.
On the other hand, the informal groups (B2, C and D) into which Marphysa has been split, have not yet been monophyletically tested. These divisions are used here to highlight the diversity of forms within the genus and help differentiate the species morphologically. However, it is imperative to carry out a species revision and a phylogenetic analysis to help reveal these artificial groups' status.
Historically, many synonyms have been subjectively proposed for Marphysa species causing a simultaneous decline in the species richness and an increase in either cosmopolitan species or species complexes. We strongly recommend that, before establishing a new synonymy, researchers should rely on the review of the type material of the species in question to describe it in detail, including the overlooked characters and to compare the species for distinguishing them accurately and ideally collecting fresh material from the type locality for molecular work and morphological variation. their support and by allowing the first author the use of installations for data analysis. To Tulio F. Villalobos-Guerrero for the comments to improve the manuscript and taxonomic advice. Thanks go to Humberto Baena-Basave (ECOSUR) for his advice on digital photography and photo editing. We are also grateful to editor Pavel Stoev and the five reviewers for their careful comments and suggestions. During this research, the first author was supported by a scholarship from CONACYT (514117/298079) and DAAD Short-Term Grants (91673476).