Tubicolous polychaete worms (Annelida) from Bahía de Chamela Islands Sanctuary, Mexico, with the description of a new bamboo worm

Abstract Background The islands and islets of Bahía de Chamela, in the Eastern Tropical Pacific, were declared as the first marine sanctuary in Mexico and has been protected since 2002. Their marine biodiversity has been documented in a series of papers in the last decade, but only three species of polychaete worms have been reported. New information Sixteen species of sedentary polychaete worms belonging to the families Maldanidae, Oweniidae, Sabellariidae, Sabellidae and Serpulidae are reported to the Bahía de Chamela Islands Sanctuary, 15 of these species constituting the first records in the area. Isocirrus tropicus (Monro, 1928) (Monro 1928) and Notaulax californica (Treadwell, 1906) (Treadwell 1906) constitute new records to Mexico; Idanthyrsus mexicanus Kirtley, 1904 (Kirtley 1994) is first recorded since its description and one species of bamboo worm (Maldanidae) is described as new to science. The new species belongs to the genus Clymenura Verril, 1900 (Verrill 1900) and its characterised by the presence of a glandular shield on chaetiger 8; a cephalic plaque oval with smooth margins and a rounded palpode; nuchal organs straight, parallel, almost full length of plaque; manubriavicular uncini present from chaetiger 1 with 3–4 teeth above the main fang without hairs or bristles; two pre-anal achaetous segments with tori; an anal funnel with alternating triangular cirri, being the longest that are located mid-ventrally.


Introduction
The islands and islets of Bahía de Chamela, in the Eastern Tropical Pacific, were declared as the first marine sanctuary in Mexico and has been protected since 2002(Miranda et al. 2011. Bahía de Chamela presents a high environmental heterogeneity, in which there is a considerable number of species (López-Uriarte et al. 2009), making it a priority site of extreme importance for the conservation of its marine ecosystems (CONABIO-CONANP-TNC-PRONATURA 2007). In addition, it is an area of fishing importance and tourist potential for the Jalisco coast.  . However, only three species of polychaete worms have been reported in Bahía de Chamela: the acoetid Polyodontes lupinus (Treadwell 1941, Pettibone 1989, the serpulid Spirobranchus sp. and the sabellid Bispira rugosa (López-Uriarte et al. 2009).
Polychaetes are segmented worms belonging to the phylum Annelida. They are predominantly marine with some species in fresh and terrestrial groundwaters Timm 2008, Glasby et al. 2009). As many as 11,456 species, 1417 genera and 85 families were recognised as valid up to 2016(Pamungkas et al. 2019. In Mexico, 1500 species have been reported along their Pacific and Atlantic littorals ). In the present contribution, 16 sedentary polychaete worms are documented to Bahía de Chamela, one of them constitutes a new record to Mexico and one bamboo worm is established as a new species to science, formally described below.

Fieldwork
The material reported in this paper was collected between April 2009 and June 2013 in Bahía de Chamela, Jalisco, Mexico (Fig. 1). The specimens were obtained by hand, snorkelling, scuba diving and using a biological dredge. Worms were fixed in 10% formaldehyde-seawater for 72 h, then excess formalin was removed, specimens were washed with tap water, left in it for 24 h to remove the remaining fixative and sea-water and later transferred to 70% ethanol for long-term preservation.

Identification
Observations and body measurements were undertaken with a Leica MZ75 stereomicroscope or an Olympus CH30 high power microscope. Photographs were taken with an attached Canon EOS Rebel T7i digital camera. Methyl green and Shirlastain-A Study area with location of sampling sites and the presence of tubicolous polychaetes.
Tubicolous polychaete worms (Annelida) from Bahía de Chamela Islands Sanctuary, ... were used for improving the contrast of surface features and analysis of the main morphological features.
The new species description is based on the holotype, with variation of paratypes as indicated in parenthesis. Except for the new species here described, descriptions are presented in a broad sense in order that any interested people can follow the identification keys with accuracy. In the nomenclature and taxon discussion sections, systematic contributions are included for those people who require a deeper analysis.

Diagnosis
A worm of 32 mm length, 1 mm width, with 14 chaetigers, two pre-anal achaetous segments with tori, manubriavicular uncini present from chaetiger 1 with 3-4 teeth above the main fang without barbules; manubriavicular uncini similar to that of chaetiger 1, but the number of barbules increases towards the most posterior segments. Cephalic plaque oval with entire, smooth margins; rounded palpode; nuchal organs straight, parallel, almost full length of plaque. No ocelli, no segmental collars. Anal funnel present with 28 distinct alternating triangular cirri: short and long (twice longer than short ones) and with the mid-ventral cirrus longest (about twice the length of the adjacent long). Chaetiger 8 glandular shield extending anteriorly and ventrally from notochaetal fascicle forming a half-oval.

Etymology
From the Latin scutata, meaning armed with a shield and referring to the glandular shield on chaetiger 8.

Taxon discussion
The specimens, here reviewed, match with the emendation of Clymenura provided by . The genus Clymenura has never been reported in Mexico. The closest record to the Mexican Pacific of a Clymenura is C. gracilis Hartman, 1969(Hartman 1969), a species originally described from Santa Monica, California. However,  considers C. gracilis as incertae sedis because the holotype lacks of ventral shield on chaetiger 8; consequently, it cannot be a member of Clymenura. Isocirrus tropicus (Monro, 1928). A) Body, anterior part, lateral view, B) cephalic plaque and first two chaetigers, lateral view, C) everted proboscis, D) anal funnel, lateral view, E) dorsal margin of cephalic plaque, F) manubriavicular uncini from 5th chaetiger, G) anal plaque, posterior view, H) mid-ventral whitish ridge starting in chaetiger 7. Scale bars: A) 3 mm, B) 2 mm, C) 1 mm, D) 1.3 mm, E) 1.8 mm, F) 400x magnification, G) 0.5 mm, H) 1.5 mm. Arrows in A pointed to lateral notches of cephalic plaque. In B, arrows pointed to crenulated margin of cephalic plaque.
it has a marked dentition of uncini in first thoracic chaetiger and 4.
it has two preanal achaetous segments (cephalic plaque notched laterally, prostomial eyes present, reduced or reminiscent dentition of uncini in first thoracic chaetiger and three pre-anal achaetous segments in C. columbiana).
Clymenura cirrata is characterised by the presence of four long anal cirri (only the midventral is long in Clymenura scutata sp. n.); the lateral margins or the cephalic plaque are notched (entire in Clymenura scutata sp. n.); and collars are present in chaetigers 2-4 (absent in Clymenura scutata sp. n.).
The rest of the species in the genus were described from high north European latitudes and the Northern Pacific region: Novaya Zemlya (one species), Japan (three species), Norway (one species), the Laptev Sea (one species), except for Clymenura snaiko Table 2.
Main features of some Clymenura species. , that was described from New Zealand (see Read 2011 to further details). Clymenura snaiko and Clymenura scutata sp. n., have the mid-ventral anal cirrus long, but the Mexican species have alternated cirri, whereas all other cirri are similar in length in C. snaiko. In addition, the lateral margins of the cephalic plaque are notched in C. snaiko versus entire in C. scutata sp. n.
Referring to intraspecific variation, a paratype was found under regeneration of the anterior end of the body (Fig. 2G). The cephalic plaque and the first three segments are a half narrower than segment 4 and those subsequent. However, the glandular shield on chaetiger 8 is present, it maintain the half-oval shape. As the number of chaetigers (and size) varies in all paratypes, their use, therefore, cannot be considered diagnostic.

Taxon discussion
Originally described from Taboga (Pacific coast of Panama) to 7-9 m depth in sand at low tide (Monro 1928). It was re-described and transferred to Isocirrus by Salazar-Vallejo (1991), based on the presence of acicular spines on chaetigers 1-3 and the presence of anal cirri of similar length. According to Salazar-Vallejo (pers. comm. 2020), the inclusion of Euclymene tropica in their key was a mistake (Salazar-Vallejo and Díaz-Díaz 2009); it must be included within Isocirrus, but having subequal anal cirri.

Taxon discussion
Detailed description and illustrations are available in Kirtley (1994).
Abdominal uncini with six pairs of teeth in side view (Fig. 7D).

Distribution
This constitutes the fist record since its establishment by Kirtley (1994).

Taxon discussion
This species was described from Bahía Tenacatita, west of the islets off Barra de Navidad, between 45.7 and 64 m depth. A second species from Western Mexico is Idanthyrsus armatopsis Fauchald, 1972Fauchald (1972

Description
Gregarious worms commonly known as "honey comb worms" (as well as I. cretus and I. mexicanus). Tubes constructed with sea shells fragments (Fig. 8A). Complete specimens 14.1-14.5 mm long, 1.3-1.5 mm wide, with 26-32 abdominal chaetigers and a caudal peduncle 4-6 mm long. Body divided into four specialised regions: operculum, parathorax (with three segments), abdomen and caudal region where there is a long peduncle (Fig. 8B). Operculum composed by a crown and a peduncle, forming two not fused lobes (Fig. 8C). Outer paleae with straight blades basally, then slightly curved towards tips (Fig. 8F); lateral denticles long, narrow and straight, those from the distal area longer than basal half ( Fig. 8F-G). Inner paleae with straight blades, transversal thecae not easily seen and pointed tips (Fig. 8H). Two nuchal hooks with limbate area in concave sides (Fig. 8E). Abdominal uncini with seven teeth in side view (Fig. 8D).

Ecology
Found in colony mixed with Idanthyrsus cretus.

Taxon discussion
Chávez-López (2019) recorded this species to Guerrero and Oaxaca and the formal description of the species is under construction (Chávez-López pers. com. 2020). Specimen from Bahía de Chamela fits the description provided by Chávez-López (2019), except for the presence of inner paleae with pointed tips versus blunt by Chávez-López (2019). Judging from her Figure 9D, we considered that the tip is pointed instead of blunt. In addition, in her Figure 10G, an inner palea from a juvenile has a blunt tip, but it seems broken.

Taxon discussion
Widely reported in the Gulf of California and Nayarit (Tovar-Hernández and Carrera-Parra 2011), this worm has been found being parasitised by Gastrodelphys dalesi, a cyclopoid copepod attached to the radioles, the inner base of the radiolar crown, dorsal lips and attached to the dorsal pockets of collar (Gómez and Tovar-Hernández 2008). A full description and illustrations are available in the revision by Tovar-Hernández and Carrera-Parra (2011). (

Description
Body 13-34 mm long, 2-4 mm wide, radiolar crown 6-17 mm long with 13-29 pairs of radioles, thorax with 13 chaetigers and abdomen with 46-82 chaetigers. Gregarious species (Fig. 10A). Soft tubes constructed by fine sand. Each radiole with three purple bands without a uniform distribution pattern (Fig. 10B). Within each purple radiolar band, there is a pair of compound eyes (three pairs of eyes per radiole). Ventral margins of thick base spiralling inwards to form three whorls (Fig. 10C). Collar with purple spots dorsally and ventrally. Thorax with purple spots anterior to each thoracic notochaetae (Fig. 10D). Purple maculae between ventral shields and tori (Fig. 10E). In lateral view, dorsal and ventral spots are seen in each segment, located at the ends of notochaetae and neurochaetae (Fig. 10F).

Taxon discussion
This is the only fan worm that has been included in the invertebrate guides from the Gulf of California (Kerstitch and Bertsch 2007). It has been reported in the Gulf of California (Kudenov 1975a) and Pacific Panama and Costa Rica (Knight-Jones and Perkins 1998).

Taxon discussion
One specimen was found regenerating a radiolar crown. This species has been reported in the Pacific coast of Panama and some localities from Mexican Pacific (Tovar-Hernández 2007).

Description
Solitary fan worm associated with dead coral and rocks. Body length 11-14 mm, width 0.9-1.1 mm. Radiolar crown length 3-3.4 mm with 8-10 pairs of radioles. Thorax with eight chaetigers and abdomen with 60-67 chaetigers. Radioles with short bands of radiolar ocelli (Fig. 12A-B), each band as long as the space of 4-6 pinnules, ocelli distributed in single rows of 12 to 16 ocelli, bands located at three quarters of the radiolar crown length (Fig. 12C). Ventral margin of collar incised, forming rounded lappets. Base of radiolar crown (basal lamina or radiolar lobes) short, as long as the length of the first three segments in lateral view.

Taxon discussion
This is the first formal record in Mexico.

Taxon discussion
Members of Parasabella seem like Acromegalomma at first view, but the latter have subdistal compound eyes in radioles (absent in Parasabella).

Taxon discussion
At first view, Pseudobranchiomma can be confused with Branchiomma because both are commonly gregarious, their bodies are full of purple maculae and both have interramal eyes and radioles with paired eyes. However, Branchiomma has long stylodes as tongue or straps-like filaments along radioles, easily seen under a stereoscope. In the Southern Gulf of California, it has been reported associated with man-made substrates in densities reaching 487 ind/m , associated often with the invasive sabellid Branchiomma bairdi (Tovar-Hernández and Dean 2014). Both species have been reported as introduced in Australia (Capa and Murray 2015, Capa and Murray 2016) and both species have been also registered in Galapagos by Keppel et al. (2019): B. bairdi as introduced, whereas P. schizogenica inside their natural distribution area. As its reproduction is mainly by architomy (fission, asexual reproduction), it is common to find small specimens or clones in a chain below the parental worm (Tovar-Hernández and Dean 2014).

Description
Body 8-10 mm long, 0.8-1 mm wide. Radiolar crown length 1.5-1.7 mm with 8-10 pairs of radioles. Thorax with seven chaetigers and abdomen with 61-64 segments. Verticil with 8-12 yellow to dark brown spines unequal in size ( Fig. 13D-H). Dorsal hook broad, curved, larger than all other spines, covering central disc. Other spines with pointed tip and pronounced knob each. First and second pair of dorsal spines (lateral to dorsal hook) with tips and trunks wider than all other spines (Fig. 13D-H). Collar chaetae bayonet with two blunt teeth; distal blade smooth.

Taxon discussion
Hydroides brachyacantha Rioja, 1941(Rioja 1941a, an important fouling serpulid species originally described from Mazatlán (Southern Gulf of California, Mexico) and Acapulco (southern Mexican Pacific), has been reported from the Mexican Pacific and numerous tropical and subtropical localities (Rioja 1941a. At the Mazatlán Port, the mean annual density of H. brachyacantha during 2009 was 80 ind m (fouling assemblages in metallic buoys), with a minimum of 4 ind m in November and a maximum of 304 ind m in March . Recently, a neotype was established by , who also demonstrated that the previous records from the species in Australia belongs to a different lineage. Salmacina tribranchiata. -Rioja 1941b: 738-739, pl. 9, figs. 11-14.-Hartman 1961: 44.-Hartman 1969: 771-772, figs. 1-6.-Salazar-Vallejo and López-Muraira 1983

Description
Tube with a prominent longitudinal ridge and a robust spine extending over the tube mouth ( Fig. 15C-D). Entire worms 26-30 mm long, 5-6 mm wide. Operculum calcareous, pinkish, oval-shaped 4 mm x 6 mm. Opercular peduncle with wide wings. Opercular plate with three short spines emerging from a short common stem, with 2-3 secondary spinules on each (Fig. 15E-G). Red radioles, in spiral arrangement with six whorls (Fig. 15A). Thoracic membrane extends to last thoracic chaetiger, forming a short ventral apron.

Taxon discussion
Widely distributed in Mexican Pacific .  reports only three wide spines on the opercular plate of S. cf. gaymardi in comparison with the five spines in specimens from Bahía de Chamela. Spirobranchus spinosus Hartman have five spines, but specimens from Bahía de Chamela cannot be attributed to that taxon, based on the following differences: in S. spinosus, the spines are thinner than in S. cf. gaymardi; each spine has 5-8 spinules versus 2-3 spinules in S. cf. gaymardi; each spines is separated one from the other, whereas spines emerge from the short common stem in S. cf. gaymardi. Consequently, the number of spines cannot be used to discern between S. cf. gaymardi and S. spinosus as suggested in some taxonomic keys.

Discussion
It is notable that the most important inventories of some of the main marine taxonomic groups in the Bay have been carried out in the last 10 years, since 2009, so that the present study provides a first approach to the polychaete tubeworm worms of Bahía de Chamela, including a new record for Mexico and the establishment of a new taxon. Lack of knowledge of Annelida, as in other marine invertebrates, is common in most of the Mexican tropical Pacific. This knowledge is particularly important in marine-protected areas, since species records are a critical tool to generate a complete fauna list as a first step to understand distribution patterns, carry out subsequent monitoring of biodiversity and finally to assess the effectiveness of protected areas. However, this information should include accurate curatorial data from sampling technique, habitat, distribution notes along with preservation and proper storage in voucher collections. The present study provided all these data and could stimulate further research on the different groups of polychaete annelids in the Bay.
Amongst the families here reported, maldanids, also known as bamboo worms, are discretely mobile, deposit feeders, inhabiting usually soft sediments (Jumars et al. 2015). Tubes of Clymenura scutata sp. n. and Isocirrus tropicus were found attached to rock surfaces, the latter being also found in sand.
Oweniids, although mostly tubicolous, are considered discretely mobile because they can extend and move their tubes within sediment. These worms are primarily surface deposit feeders, but some species with tentacular crowns also suspension feed (Jumars et al. 2015). Owenia collaris have tentacular crowns and were found in sand.
Sabellariids, also known as honeycomb, sand-castle or sand-mason worms, usually build their tubes either attached to hard substrata or to other sabellariid tubes, in some cases forming massive reefs, but some species are solitary (Jumars et al. 2015, Helm et al. 2018). In Bahía de Chamela, Idanthyrsus cretus and I. mexicanus were found attached to rocks and corals, while the potential new species of Idanthyrsus (here refered to as Idanthyrsus sp.) after Chávez-López (2019)) was found amongst tubes of I. cretus.
Sabellids, known as feather-duster worms, fan worms or sea flowers, are sessile, tube builders. They inhabit tubes that they build with secreted mucus and attached mud or sand particles, except Glomerula that builds a calcareous tube (Capa et al. 2019 Introduced species were not detected in Bahía de Chamela, but their presence cannot be discarded since all specimens reviewed in the present study proceed from natural substrates. Consequently, a specific monitoring programme is desirable for the early detection of harmful tubicolous worms in that natural protected area.