Molecular analysis of Indo-Pacific Syllinae annelids with the description of five new species from the Philippine Islands

Syllinae is the largest and the most diverse subfamily in Syllidae and usually a dominant group in many marine habitats, both in terms of numbers of species and individuals. However, the lack of morphological synapomorphies for many genera and the non-monophyly of most of these suggested by molecular studies, have stressed taxonomic and systematic problems within this subfamily. Although several studies have revised some genera in the Indo-Pacific area, important gaps of knowledge remain for this biogeographic region. Thus, in this study we undertook detailed morphological examination of Syllis and Megasyllis specimens from the Philippine Islands and provide a phylogenetic analysis of Indo-Pacific Syllinae in order to evaluate the position of newly described species. We provide line drawings, light microscopy observations and electron microscopy micrographs for eight species, including five new to science. Maximum likelihood and Bayesian inference analyses of four molecular markers (28S rRNA, 18S rRNA, 16S rRNA, cytochrome c oxidase subunit I) from 85 specimens are presented, showing the non-monophyletic status of Syllis and Megasyllis. Overall, our study increases the biodiversity of syllids in this poorly known Indo-Pacific area, highlighting the importance of examining both morphological and molecular data to expand our knowledge of the subfamily Syllinae and to further resolve the taxonomic issues that remain in this group. The following new taxa are described: Megasyllis kurui sp. nov. Moreno-Martínez, San Martín & Álvarez-Campos, Syllis ireneae sp. nov. Moreno-Martínez, San Martín & Álvarez-Campos, Syllis santii sp. nov. Moreno-Martínez, San Martín & Álvarez-Campos, Syllis tini sp. nov. Moreno-Martínez, Syllis walong sp. nov. Moreno-Martínez, San Martín & Álvarez-Campos. http://zoobank.org/urn:lsid:zoobank.org:pub:E35C528D-E02D-4735-B62C-2EF5C64BBCAF


Introduction
Syllinae Grube, 1850 is the most abundant and speciose subfamily of the family Syllidae Grube, 1850, with $407 valid species in 24 genera (WoRMS Editorial Board, 2022).In the Pacific and Indian Oceans the most diverse genera of Syllinae are the type genus, Syllis Savigny in Lamark, 1818, with around 100 species of a total of 175 distributed in this region, and Megasyllis San Mart ın, Hutchings & Aguado, 2008, with all except one of its 15 species restricted to the Indo-Pacific area (e.g., Aguado & Glasby, 2015;Alvarez-Campos et al., 2015a, 1965b;Alvarez-Campos & Verdes, 2017;Hartman, 1965;San Mart ın et al., 2014, 2017a, 2023).The taxonomy and systematics of these two genera require a thorough revision since all the phylogenetic analyses published to date showed both genera to be non-monophyletic (e.g., Aguado et al., 2012, Aguado & Glasby, 2015;Alvarez-Campos et al., 2015b;Ba-Akdah et al., 2018).Syllis has been traditionally considered a genus without clear morphological synapomorphies where lots of species are dumped, and although there have been some attempts to reorganize the genus (see details in Alvarez-Campos et al., 2015b), the lack of molecular data for most species, including its type, Syllis monilaris Savigny in Lamarck, 1818, does not allow this question to be satisfactorily resolved ( Alvarez-Campos et al., 2015a).On the other hand, Megasyllis, erected by San Mart ın et al. (2008) to include some species previously identified within Syllis, Eusyllis Malgrem, 1867, and the currently non-valid genus Typosyllis Langerhans, 1879, is well defined morphologically by the presence of segments with multiple annuli, posterior dorsal cirri that appear to be pseudoarticulated, and an opaque body with a very variable size (from just a few to more than 30 mm) (San Mart ın et al., 2008, 2014).However, although these characters were deemed informative for describing Megasyllis as a new genus, all current molecular phylogenies reject its monophyly, as Megasyllis species appear together with species of the genus Alcyonosyllis Glasby & Watson, 2001(Aguado et al., 2012;Aguado & Glasby, 2015;Aguado & San Mart ın, 2009;Ba-Akdah et al., 2018).Given that the morphological synapomorphies shared by these two genera are not clear (see Aguado & Glasby, 2015 for details) and since the most conspicuous character to delimit syllid species (i.e., chaetal morphology) is not shared by the members assigned to these two genera, Megasyllis and Alcyonosyllis are still considered as separate genera in the syllid literature.
In the Philippine Islands, Grube (1878) reported several Syllis and Typosyllis species that were later synonymized with other species or considered nomina dubia (Licher, 1999).With respect to Megasyllis, the only species described from the Philippines is M. chrissyae San Mart ın et al., 2014.In fact, there have been only a handful of recent studies on the biodiversity of both genera in the Philippines (Mart ınez & San Mart ın, 2020;San Mart ın et al., 2014, 2021) and, therefore, detailed research on Syllidae is still required for this area.Thus, the present study intends to contribute to the knowledge of syllid biodiversity in the Indo-Pacific region and specifically in the Philippines, describing in detail eight species of Syllis and Megasyllis, including five new to science.A phylogenetic hypothesis is also provided for the whole subfamily using a multilocus approach of four molecular markers from 85 specimens.

Sampling and morphological examination
Specimens were collected during a biological survey in December 2010, by hand, snorkelling, or scuba diving, on algae, sponges, and coral rubble from intertidal and subtidal zones in five localities in Luzon (Republic of the Philippines).The material examined was collected within the framework of the project Systematic and taxonomic characterization of the Syllidae (Polychaeta) family based on molecular and morphological data.The problem of cosmopolitan species and Biodiversity in the Pacific (CGL2009-12292 BOS, Ministerio de Ciencia e Innovaci on, Spanish Government) for which all necessary permits were sought from the Philippine Government through the National Museum of the Philippines and the Municipality of El Nido.All the specimens were sorted in the field using a Nikon SMZ-1 stereomicroscope and then fixed and preserved in 96% EtOH, for morphological and molecular analyses.Further examination and identification were completed under two compound microscopes, an Olympus CX43 and a Nikon Optiphot, the latter with a differential interference contrast system (Nomarsky), at the Universidad Aut onoma de Madrid (UAM).Drawings were made to scale with a camera lucida attached to the Nikon Optiphot microscope.The width of specimens was measured at the level of the proventricle, excluding the parapodia.The number of segments on specimens were provided excluding first segment and pygidium, as done traditionally in the annelid literature, as these were once considered pseudosegments.For scanning electron microscopy (SEM), selected specimens were prepared on an Emitech K850 Critical Point Dryer, gold-coated with a Q150T-S Turbo-Pumper Sputter Coater and examined with a Hitachi S-3000N SEM at the Servicio Interdepartamental de Investigaci on (SIDI) of the UAM.
All the newly collected specimens were deposited in the Museo Nacional de Ciencias Naturales, Madrid (MNCN) and in the Museum of Comparative Zoology, Harvard University, Cambridge, MA (MCZ).Catalogue numbers, locality information, coordinates, substrates, and collecting dates are provided in Table 1.

DNA extraction and PCR amplification
Molecular markers used for the analyses consisted of the nuclear genes 18S rRNA ($1800 bp) and a fragment of 28S rRNA (351 bp) and fragments of the mitochondrial 16S rRNA (470 bp) and cytochrome c oxidase subunit I (COI, $650 bp).Genomic DNA was extracted from 2-3 segments of at least one individual per species, using the Speedtools DNA kit (Biotools) or the DNeasy Blood & Tissue Kit (Qiagen) following manufacturer's instructions.The 18S rRNA was amplified in three overlapping fragments using primer pairs 18S1F-18S3R, 18S3F-18Sbi, and 18Sa2.0-18S9R(Giribet et al., 1996;Whiting et al., 1997).Primers 28Sa and 28Srd5b (Edgecombe & Giribet, 2006) were used to amplify the 28S rRNA fragment.Primers 16SarL and 16SbrH (Palumbi, 1996) were employed to amplify 16S rRNA and the modified primers with inosine

Syllis marugani
Manazuru Peninsula, Japan  jgLCO1490 and jgHCO2198 (Geller et al., 2013) were used for COI.Polymerase chain reactions (PCR) consisted of 1.5 mL of DNA template in 13 mL reaction volumes containing 0.5 mL of each of 10 mM primers, and 12 mL of 1.25 U/mL REDtaq DNA Polymerase (Sigma).The temperature profiles to amplify all the markers were the same as those described in Alvarez-Campos et al. (2017a, 2017b).Using the forward and reverse primers described above, 10 mL of the PCR product was used for sequencing at the Bauer Core (Harvard University) or at the Servicio de Secuenciaci on Sanger, Unidad de Gen omica (Universidad Complutense de Madrid).
Sequences were edited in GENEIOUS 6.1.8(Kearse et al., 2012); primers were removed and overlapping fragments of the 18S rRNA marker were merged into a contig sequence.Each of the four genes was aligned in the online server of MAFFT under default parameters (Katoh et al., 2019) and subsequently concatenated with SeaView (Galtier et al., 1996).

Phylogenetic analyses
In order to assess the phylogenetic relationships of all the newly collected species and its closely related taxa, we analysed the 4 molecular markers for a total of 85 individuals in 65 putative species of Syllinae.(2), Plakosyllis Hartmann-Schr€ oder, 1956 (1), Syllis (50), and Trypanosyllis Clapar ede, 1864 (2), as well as sequences of 4 species of Perkinsyllis San Mart ın, L opez & Aguado, 2009 as outgroups (Fig. 16, Table 1).All mitochondrial and nuclear data sets were concatenated, but model selection kept each marker as a separate partition in RAxMLGUI v. 2.0.7 (Edler et al., 2021).The best-fitting model of sequence evolution was selected using the Akaike information criterion (AIC) in jModeltest 2 (Darriba et al., 2012;Guindon & Gascuel, 2003).The most appropriate partitioned model for the data set of both mitochondrial and nuclear markers was a general time-reversible (GTR) model of sequence evolution with gamma-distributed rates across sites and a proportion of invariable sites (GTR þ G þ I).Maximum likelihood (ML) analyses were run in RAxMLGUI v. 2.0.7 (Edler et al., 2021) Bootstrap support values were estimated using 500 replicates and 10 initial trees (Stamatakis et al., 2008).Bayesian inference (BI) analyses were also run with the concatenated data set using the GTR þ G þ I model, implemented in MrBayes 3.2.1 (Ronquist et al., 2012), with four Markov chains starting from a random tree that was run simultaneously for 20 million generations, with trees sampled every 2000 generations (samplefreq ¼ 2000).The initial 25% of the trees were discarded (burninfrac ¼ 0.25) after evaluating convergence with Tracer v1.6 (Rambaut et al., 2014).The resulting phylogenetic trees were visualized in FigTree v1.4.2 (Rambaut et al., 2014).Zoobank ID: urn:lsid:zoobank.org:act:51E4CAB2-6FE6-4A90-92D9-260FFE5FA12C
Midbody and posterior cirri without visible articulation.
Compound bidentate chaetae similar all along the body, with proximal and distal teeth similar in length, and short spines on margin .Posterior aciculae distally bent (Fig. 2g).
Remarks.In regard to the colour pattern, the most similar species to Syllis santii sp.nov.are S. lunaris (Imajima, 1966)  Etymology.The new species is named after Santiago G omez-Escabia, 'Santi', colleague and friend, for his invaluable help and support during the time shared in the lab.

Distribution.
Known only from the type locality.9e).Dorsal and ventral simple chaetae absent.
Remarks. ).On the other hand, S. santii sp.nov.and S. tini sp.nov.differ in the colour pattern since the former presents a thick transversal darkish-red band on each segment (Fig. 1c) while S. tini sp.nov.has also a transversal band in anterior segments with an unpigmented square in the middle on them (Fig. 1d).Syllis guidae differs in the shape and length of cirri, which are thicker and shorter than those in S. tini sp.nov., and also in the chaetae that present blades with all the spines on margin very large (Nogueira & Yunda-Guarin, 2008).Finally, S. villenai presents dorsal cirri much longer than the new species, and the chaetae with large spines on margin of the blades, surpassing the proximal tooth (Aguado et al., 2008).
Etymology.The new species is named in memory of my beloved father, Agust ın Moreno, and his nickname "Tini".He was the most joyful and generous person I have ever met.He taught me to enjoy every moment in my life and he always encouraged me to follow all my dreams with perseverance and passion.Agust ın Moreno passed away during the development of this project.Batangas Bay, 'Sepok Point ' (13.68, 120.855556), coral rubble, 6 m, 10 December 2010; one specimen in 96% EtOH (MNCN 16.01/19181), Luzon Island, between Balayan Bay and Batangas Bay, 'Mainif Point' (13.68, 120.855556), coral rubble, 1-2 m, 8 December 2010.All samples collected by G. San Mart ın Project CGL2009-12292 BOS collecting team.
Remarks.Syllis walong sp.nov.has a characteristic colouration with an eight-shaped pattern in anterior segments that is also present in other species such as S. variegata Grube, 1860 from the Mediterranean Sea, Australia, and the East Atlantic Ocean, S. ferrani Al os & San Mart ın, 1987 from the Mediterranean Sea, S. alosae San Mart ın, 1992 from Cuba, and S. nigropunctata from Australia (Al os & San Mart ın, 1987;Grube, 1860;Haswell, 1886;San Mart ın, 1992).However, anterior blades of S. walong sp.nov.are much longer than those in S. variegata and S. nigropunctata, and posterior shafts do not present the characteristic spur of the new species (Haswell, 1886;San Mart ın, 2003).In addition, S. ferrani and S. alosae differs from S. walong sp.nov. in the length of dorsal cirri, which are shorter than those in S. walong sp.nov.(Al os & San Mart ın, 1987;San Mart ın, 1992).Moreover, S. alosae presents longer blades in all chaetae with larger spines on margin, and S. ferrani has dorsal simple bidentate chaetae with some few spines on margin, not present in S. walong sp.nov.(Al os & San Mart ın, 1987;San Mart ın, 1992).On the other hand, the new species presents similar long dorsal cirri and chaetae to those in S. maganda from the Philippines and S. cambuk Mart ınez & San Mart ın, 2020 from East Timor (Mart ınez & San Mart ın, 2020).However, midbody and posterior shafts are much thicker in the new species, with a pronounced spur not present in S. maganda and S. cambuk (Mart ınez & San Mart ın, 2020).
Remarks.Syllis sp. 1 is similar to S. komodoensis Aguado, San Mart ın & ten Hove, 2008 from Indonesia, S. hyalina Grube, 1863 from Croatia, S. armillaris (O.F.M€ uller, 1776) from Denmark, S. augeneri Haswell, 1920, S. kai San Mart ın et al., 2021 from Madagascar, and S. lutea (Aguado et al., 2008;Hartmann-Schr€ oder, 1960;San Mart ın et al., 2021, 2023), in the slender shape of their bodies, with dorsal cirri similar in length to the body width (Aguado et al., 2008;San Mart ın et al., 2021;2023), except for S. lutea that has much longer dorsal cirri (Hartmann-Schr€ oder, 1960).In addition, they all present similar chaetae, with bidentate blades and teeth of the same length, with the exception of S. kai and S. hyalina that have the proximal tooth much shorter than the distal one.However, all these species show several different characters that lead us to consider Syllis sp. 1 as a different species.Thus, all those similar species present very different aciculae and unidentate ventral chaetae that are absent in Syllis sp. 1 (Aguado et al., 2008;San Mart ın et al., 2021, 2023).Syllis komodoensis, S. augeneri, and S. lutea also present shafts with distal spines, that were not seen in Syllis sp. 1. Syllis komodoensis shows some colouration in the dorsal cirri, S. augeneri has longer anterior blades than those in the Philippine species (Aguado et al., 2008) and S. lutea also differs in the length of spines of compound chaetae that reach the proximal tooth (Hartmann-Schr€ oder, 1960).Syllis hyalina, S. armillaris, and S. kai present much shorter blades in all chaetae, especially in the midbody and posterior ones.Furthermore, all of them have a shorter pharynx and S. kai also presents a shorter proventricle as well as unidentate ventral chaetae not found in Syllis sp. 1 (San Mart ın, 2003;San Mart ın et al., 2023).Although we think that Syllis sp. 1 might also be a new species, we cannot formally describe it herein given the paucity of specimens.The fact that only one individual is available does not allow us to properly show detailed morphological characters for the species since we could not examine it through SEM, and could not take into account possible intraspecific differences.We consider that describing the morphology of this unknown species might be useful for future studies in the Indo-Pacific area.Description.Specimen incomplete, lacking most posterior segments, mm long, 0.3 mm wide, with 42 segments.Body slender, elongated, without colour pattern.Prostomium oval with two pairs of red eyes in trapezoidal arrangement (Fig. 14a).Median antenna inserted in middle of prostomium, broken; lateral antennae, with 20-23 articles.Palps elongated, 1.1Â longer than prostomium (Fig. 14a).First segment slightly shorter than subsequent segments.Dorsal tentacular cirri with about 30 articles, ventral ones shorter, with 18-20 articles.Dorsal cirri slightly longer than body width, with 24-26 articles (Fig. 14a).Ventral parapodial cirri shorter to parapodial lobes.Anterior parapodia each with 6-8 compound chaetae; bidentate blades, both teeth similar, or proximal tooth slightly larger than distal one, large spines on margin, reaching level of proximal tooth (Fig. 14b); most dorsal chaetae, 1-2 per parapodium, distinctly longer (31 lm) than ventral ones (17 lm) (Fig. 14b).Midbody parapodia each with 5-6 compound chaetae, bidentate blades, shorter than those of anterior parapodia, almost all similar in length (14 lm dorsalmost, to 11 lm ventralmost), both teeth similar or proximal tooth slightly larger than distal one, with large spines on margin; 2 distal spines longer, reaching proximal tooth (Fig. 14c).Posterior parapodia each with 4-5 compound chaetae, all blades similar in length (14 lm dorsalmost, to 13 lm ventralmost), proximal tooth slightly larger than distal one (Fig. 14d); two or three distal marginal spines longer than remaining, reaching proximal tooth (Fig. 14d).Dorsal and simple chaetae absent.Posterior parapodia with single acicula each, thick, acutely pointed (Fig. 14e).Pharynx long, extending through about 13 segments (Fig. 14a).Proventricle through about 7 segments, with about 31 rows of muscle cells (Fig. 14a).Pygidium unknown.Description.Specimen 5 mm long, 0.5 mm wide, 144 chaetigers.Body relatively short and wide, somewhat dorso-ventrally flattened (Fig. 15a), yellowish with three brown transverse stripes across anterior and midbody segments; median stripe slightly thicker and two stripes thinner, in both anterior and posterior margins of segment (Fig. 15a).Prostomium oval with two pairs of red eyes in trapezoidal arrangement.Palps oval, similar in length to prostomium, fused at bases.Antennae originating on anterior margin of prostomium, median antenna with about 33 articles; lateral antennae slightly shorter, with 23-25 articles (Fig. 15a).Dorsal tentacular cirri slightly shorter than antennae, with about 24-27 articles; ventral tentacular cirri shorter, with about 18-20 articles (Fig. 15a).First segment slender, distinctly shorter than subsequent segments, dorsally reduced.Anterior dorsal cirri longer than body width, with 23-28 articles, midbody dorsal cirri with 20-25 articles, and posterior dorsal cirri with 15-17 articles.Ventral digitiform cirri, shorter than parapodia.Compound chaetae bidentate.

Remarks
All chaetae similar throughout body, with both teeth similar in length or proximal tooth slightly shorter than distal one, and short spines on margin (Fig. 15b-d).
Remarks.Syllis sp. 3 is similar to the recently described S. thylacine San Mart ın et al., 2023 from Australia in the colour pattern and chaetae (San Mart ın et al., 2023).Thus, they both share the transverse stripes across anterior and midbody segments, although S. thylacine also presents some lateral spots not seen in Syllis sp. 3. Compound bidentate chaetae with similar length of both teeth are present in both species, but spines on margin are larger in S. thylacine.Midbody and posterior simple chaetae are also similar but in Syllis sp. 3 they present fewer and shorter spines than in the Australian species (San Mart ın et al., 2023).On the other hand, Syllis sp. 3 is very similar in the general aspect of the body to the original description of Syllis zonata (Haswell, 1886), since it presented a thick body and cirri, and similar colour pattern.Furthermore, they both present very similar compound bidentate chaetae, bidentate, with both teeth similar in length or proximal tooth slightly shorter than distal one, and short spines on the margin (Augener, 1913;Haswell, 1886).Syllis zonata was later synonymized to S. prolifera (Krohn, 1852) by Licher (1999), although they present several differences in cirri, chaetae, and aciculae (Campoy, 1982;Fauvel, 1923;Krohn, 1852;San Mart ın, 2003).Thus, both Syllis sp. 3 and S. zonata have thicker cirri, short spines in the margin and very rounded aciculae, not seen in S. prolifera.Thus, we suggest that our specimen could be S. zonata, that would be considered again a valid species.However, since the type material of this species is lost and there is only a midbody fragment of a specimen from Australia, an accurate comparison and identification with the Philippine material has not been possible yet.

Discussion
The results obtained in our study contribute to the knowledge of the biodiversity of the subfamily Syllinae, increasing the number of species in the group.With the four newly described species of Syllis, the genus now comprises a total of 179 described valid species, whereas Megasyllis contains 16 species.We provide detailed morphological descriptions for the five new described species of Megasyllis and Syllis and for the three Syllis species that could not be named because we only collected one specimen.In addition, we provide molecular information for all of them, except for Megasyllis kurui sp.nov.since we could not amplify any molecular marker.Furthermore, we include sequences of three species from the Indo-Pacific region, Alcyonosyllis aidae, Megasyllis eduardoi and Paraopisthosyllis alternocirra, and expand the geographic distribution of these two latter species (Table 1).All the newly described species of Syllis were found within a large clade, here named clade 4, that includes other species from the genus together with those of Paraopisthosyllis, Megasyllis, and Alcyonosyllis,  although the internal relationships of the clade were poorly supported.
As was already reported in previous studies, both our ML and BI analyses also showed the chaotic state of the phylogenetic relationships within the subfamily Syllinae, since most of the genera it contains were nonmonophyletic (Aguado et al., 2012;Aguado & Glasby, 2015;Alvarez-Campos et al., 2015b;Ba-Akdah et al., 2018).In our analysis, the genus Syllis was clearly paraphyletic and divided in either two or four main clades that nested with other Syllinae genera such as Trypanosyllis, Plakosyllis, Eurysyllis, Haplosyllis, Branchiosyllis, Paraopisthosyllis, Megasyllis, and Alcyonosyllis (Fig. 16 and Suppl.Mat. 1).These results partially agree with previous phylogenetic studies in Syllis from Australia and the Red Sea, where the genus also appeared to be divided in four main clades ( Alvarez-Campos et al., 2015b;Ba-Akdah et al., 2018).Some general discrepancies were found between our analysis and these previous studies, probably due to the differences in the number of specimens included, as well as in the number of molecular markers analysed ( Alvarez-Campos et al., 2015b;Ba-Akdah et al., 2018).Interestingly, some of the clades we have recovered here include species that seem to share some morphological features, such as the type of chaetae, aciculae or cirri (clades 1, 2, and 3; Fig. 16).Clade 1 grouped species that belong to the S. gracilis species complex (see Alvarez-Campos et al., 2017;Langeneck et al., 2020), i.e., which present short dorsal cirri on midbody and posterior segments, and the reduction of blades on posterior chaetae becoming unidentate and in some cases ypsiloid-shaped.Clades with similar species were found in previous phylogenetic studies, such as clade II in Alvarez-Campos et al. (2015b) or clade B in Ba-Akdah et al. (2018), indicating that the presence of this type of chaetae might bear a phylogenetic signal.The Syllis species included in our clade 2 present a slender body, chaetae with 1-3 dorsal blades distinctly longer than the ventral ones; the bidentate blades present the proximal tooth slightly longer than the distal one and the spines on margin are usually long, reaching the level of the proximal tooth (Figs 13 and 14; see also Hartmann-Schr€ oder, 1960).Clade 2 also corresponds to the previously reported clade I in Alvarez-Campos et al. (2015b) and clade C in Ba-Akdah et al. (2018).However, in these two cases, more species with similar chaetae were included and were well supported in the clade, such as S. yallingupensis (Hartmann-Schr€ oder, 1982) and S. broomensis (Hartmann-Schr€ oder, 1979), but this grouping was not supported in our analyses (Fig. 16 and Suppl.Mat. 1).Our clade 3 includes species with anterior chaetae with long blades and midbody and posterior chaetae with 1-3 dorsal blades longer than ventral ones (Licher, 1999); furthermore, the most basal spines on margin of all chaetae are longer than distal ones and the posterior parapodia present at least one acicula distally curved.In particular, the sequenced specimens of S. walong sp.nov.appeared closely related to the species S. heronislandensis (subclade II; Fig. 16), although these two species do not seem to share any obvious morphological features.In fact, S. heronislandensis does not present any characteristic colour pattern, it has short dorsal cirri, and the anterior chaetae show ventral spines longer than the distal ones (Hartmann-Schr€ oder, 1991).In contrast, S. walong sp.nov., presents a striking colouration, with very long cirri, and the spines on the chaetal margin being all of the same length.Nevertheless, we could not verify the identity of S. heronislandensis specimen since its location was not mentioned in the NCBI database, and therefore, we cannot be sure whether this was a misidentification of a similar species.Our clade 3 corresponds to clade III in Alvarez-Campos et al. (2015b) and clade A in Ba-Akdah et al. (2018), showing again that this specific detail in chaetae might provide some phylogenetic signal.However, the molecular information currently available for species and specimens sharing the morphological characters of each clade is still insufficient to propose robust reassignments.
On the other hand, and in addition to the common morphological features mentioned above, some of the clades we obtained also showed similarities in the type of stolon their species produce during reproduction (clades 1, 3, and 4; Fig. 16), although it is worth noting that this character has not been reported in many species of each Syllinae clade (see asterisks indicating it in Fig. 16).Thus, clade 1 grouped species with pentacerous stolons, that have a well-developed prostomium, two pairs of eyes and three antennae; clade 3 included species with tetracerous stolons that are those with two pairs of eyes and one pair of well-developed articulated antennae; and clade 4 comprised a mixture of species that do not have any characters in common except for the presence of dicerous stolons, which present two pairs of eyes and one pair of small unarticulated antennae (Fig. 16 and see similarities with clades obtained in Alvarez-Campos et al., 2015b).The hypothesis of the type of stolon as an important evolutionary character within the subfamily, was previously suggested in a study of the whole family Syllidae, where the different clades obtained in Syllinae were defined by the different types of stolon (Aguado et al., 2012).However, our study does not fully support this hypothesis, given that only three of the four clades present similarities in the stolon morphology, and actually, in most cases this is only Indo-Pacific Syllinae annelids from the Philippine Islands based on the few species for which the reproductive mode has been reported only in one species for the clade 3).
It is also important to mention that clade 4 also included species from the genera Megasyllis, Alcyonosyllis, and Paraopisthosyllis, in agreement with the results obtained in previous phylogenetic studies (e.g., Aguado & Glasby, 2015;Ba-Akdah et al., 2018).In both our analyses, Megasyllis and Alcyonosyllis have resulted to be paraphyletic, in contrast to previous ones where Alcyonosyllis was still considered a monophyletic group (Aguado & Glasby, 2015).Our results showed a relatively well-supported clade (BS ¼ 94%; PP ¼ 1) that contained both Megasyllis and Alcyonosyllis species and, thus, the solution to reorganize their taxonomy would be to consider Megasyllis a junior synonym of Alcyonosyllis, as discussed earlier (Glasby & Watson, 2001).However, given the lack of clear morphological synapomorphies to define this clade, that only seem to have in common the shape and arrangement of dorsal cirri and the body with distinct colour patterns, it is preferable to maintain both genera until more material and sequences can be investigated in a broader analysis, including the newly described species in the present paper.With the current lack of information for most part of Syllinae species, and specifically for the type species, Syllis monilaris, which has not yet been sequenced, it is not possible to reorganize this complex subfamily and, therefore, a detailed morphological study combined with all the available molecular data is still required in order to delimit the status of the type genus Syllis as well as other non-monophyletic genera, such as Megasyllis, to properly establish the phylogenetic relationships within Syllinae.
Adequately resolving the taxonomy and systematics of highly diverse groups, such as syllid worms, might have an impact in conservation of marine ecosystems.Specifically, cataloguing species in those geographical areas poorly prospected but with great marine diversity, as the Indo-Pacific region, is an urgent task before these habitats are irreversibly transformed or even disappear.Although in the past few years research on marine invertebrates has increased in the Philippines, most has focused on groups with commercial value, such as molluscs, crustaceans, or echinoderms (Bautista et al., 2017).However, as our results have shown, a vast number of species of several other dominant taxa, such as annelid worms, still remain undiscovered in this area.Thus, more taxonomic studies are required in such neglected regions in order to provide a comprehensive framework of all the extant ocean biodiversity for its correct management.

Conclusions
In the present study, we have increased the number of species from the subfamily Syllinae and, specifically, from the genera Syllis and Megasyllis, which seem to be the most abundant genera in the Indo-Pacific region.We have also shown that the current organization of the genera Syllis and Megasyllis is not phylogenetically informative since they are both paraphyletic groups.We are currently developing further studies involving more species and specimens, as well as combining molecular and morphological data within the subfamily Syllinae in order to clarify the relationships within it and to resolve the systematics of its non-monophyletic genera.We have also highlighted the importance to study highly diverse marine taxa in remote geographic areas where the existing diversity is largely unknown.That can result in the proper management of marine ecosystems and contribute to their conservation.

Fig. 16 .
Fig. 16.Phylogenetic relationships of the new sequenced species from the Philippines with the rest of Syllinae, inferred from the Maximum-likelihood analysis of the four concatenated molecular markers (18S rRNA, 28S rRNA 16S rRNA, and COI).Numbers above branches indicate bootstrap support values (only BS !75% are indicated).Numbers below branches indicate posterior probabilities support values (only PP !0.95 are indicated).Asterisks show the species with known type of reproductive stolon.Drawings showing the morphology of the different supported clades resulting from the analyses (modified from San Mart ın, 2003; Alvarez-Campos et al., 2013).

Table 1 .
Localities, substrates, date of sampling, coordinates, catalogue numbers, and GenBank accession numbers for all specimens included in the analyses.All the new generated sequences are in bold.
(Hartmann-Schr€ oder, 1984)Schr€ oder, 1984), and M. tigrina San Mart ın et al., 2014, all from Australia, to M. eduardoi San Mart ın et al., 2014 from New Zealand, to M. nipponica (Imajima, 1966) from Japan, to M. chiki San Mart ın et al., 2021 from China, and to M. chrissyae San Mart ın et al., 2014 from the Philippines Megasyllis eduardoi and M. kurui sp.nov.are also similar in the robustness of their bodies and in having well-articulated cirri, but M. eduardoi is much larger than the new species and it does not have simple chaetae in midbody parapodia.
(San Mart ın et al., 2014)cans also differs from M. kurui sp.nov.on its body, because it is larger and thicker and it presents non-articulated dorsal cirri(San Mart ın et al., 2008).On the other hand, M. tigrina, M. chiki, and M. subantennata present blades of compound chaetae much longer than those in M. kurui sp.nov.withproximal and distal teeth more separated(San Mart ın et al., 2014).Moreover, the dorsal simple chaetae of M. subantennata do not have spines on margin.Although M. glandulosa and M. kurui sp.nov.presentsimilarcolouration in the prostomium, they differ in the presence of midbody unidentate simple chaetae absent in the new species(San Mart ın et al., 2014).Etymology.The new species is named after Joan Corominas' nickname ('kuru'), the first author's cousin, for his unwavering support, love, and passion for the sea life.
(Figs 6a, 7a).Median antenna arising in middle of prostomium, with 28 articles; lateral antennae 1.17Â shorter than median one, with about 24 articles (Figs 6a, (San Mart ın et al., 2023)2008;N uñez & San Mart ın, 1991) 2008;N uñez & San Mart ın, 1991;San Mart ın et al., 2023)nt dorsal dark bands on each segment(San Mart ın et al., 2017a).However, in S. warrnamboolensis and S. noolinga these bands are only in anterior segments, and in S. lunaris there are two bands per segment instead of one(San Mart ın et al., 2017a;2023).In addition, they all differ in the morphology of compound chateae.In this sense, posterior compound chaetae in the new species are similar to those in S. yallingupensis(Hartmann-Schr€ oder, 1982)from Australia, S. lutea (Hartmann-Schr€ oder, 1960) from the Red Sea, S. guidae Nogueira & Yunda-Guarin, 2008 from Brazil, and S. cruzi N uñez & San Mart ın, 1991 and S. profunda Cognetti, 1955 from the Mediterranean Sea and Australia, since all are bidentate with proximal tooth larger than distal one(Hartmann- Schr€ oder, 1960, 1982;Nogueira & Yunda-Guarin, 2008;N uñez & San Mart ın, 1991;San Mart ın et al., 2023).However, S. yallingupensis presents the most distal spines on margin much longer than the ones observed in Syllis santii sp.nov., and it also presents spiniger-like chaetae, which are absent in the new species(San Mart ın et al., 2017a).Syllis lutea differs in anterior and midbody aciculae, since they are thick and acuminate(San Mart ın et al., 2017a)in contrast to the thin and distally bent ones observed in the new species.On the other hand, blades of all chaetae in S. guidae and S. cruzi are shorter than those in the new species, with longer spines on margin that reach the proximal tooth in midbody and posterior parapodia(Nogueira & Yunda- Guarin, 2008).In addition, they both also differ in dorsal cirri, since they are shorter and thicker than the ones present in the new species(Nogueira & Yunda-Guarin, 2008;N uñez & San Mart ın, 1991).Finally, S. profunda presents thick shafts with a subdistal protuberance in posterior parapodia not present in Syllis santii sp.nov.(SanMartın et al., 2023).
(Aguado et al., 2008;ies to S. tini sp.nov.are the previously mentioned S. santii sp.nov., S. cruzi, and S. guidae, as well as S. villenai Aguado, San Mart ın & ten Hove, 2008 from Indonesia.All of them have posterior bidentate compound chaetae with the proximal tooth larger than the distal one(Aguado et al., 2008; (N uñez & San Mart ın, 1991;Alvarez-Campos et al., 2015b;San Mart ın et al., 2017b).In addition, blades in S. santii sp.nov.and S. cruzi have distal spines longer than the rest, reaching proximal tooth, as it occurs in the new species.However, in S. santii sp.nov.thesedistal spines only appear in midbody and posterior chaetae and S. cruzi presents simple bidentate chaetae not observed in S. tini sp.nov.(Nuñez & San Mart ın, 1991;Alvarez-Campos et al., 2015b . Syllis sp. 2 is similar to S. hampirmenyatu Mart ınez & San Mart ın, 2020 from East Timor, S. dominguezi San Mart ın et al., 2021 from China, and S. terraeignium Soto et al., 2020 from Chile in the slender body shape, without colouration, with all cirri similar in length to the body width, and a very long pharynx (Mart ınez & San Mart ın, 2020; San Mart ın et al., 2021; Soto et al., 2020).Nevertheless, they all differ in the morphology of the bidentate chaetae with proximal tooth shorter than the distal one and in the number of aciculae more numerous than in Syllis sp. 2 (Mart ınez & San Mart ın, 2020; San Mart ın et al., 2021; Soto et al., 2020).In addition, S. terraeignium and S. dominguezi present dorsal and ventral chaetae not seen in Syllis sp. 2 and S. dominguezi is the only one with chaetae with long spines on margin but without the most distal ones reaching the proximal tooth as occurs in Syllis sp. 2 (San Mart ın et al., 2021).Posterior chaetae in S. hampirmenyatu have very reduced blades and robust shafts that are not seen in Syllis sp. 2. All these differences lead us to consider Syllis sp. 2 as a new species but cannot formally describe it herein given the paucity of specimens, as discussed for the previous species.