The annelid family Flabelligeridae is generally recognized as a sedentary group living within soft sediments or hard substrates (Rouse and Pleijel 2001); however, recent morphological and molecular phylogenetic studies show that the family also contains some holopelagic species (Burnette et al. 2005; Osborn and Rouse 2008; Salazar-Vallejo et al. 2008). In Japan, the records of flabelligerids are relatively few and scattered. To our knowledge, the following 13 species of Flabelligeridae have so far been recorded from Japanese waters: Brada inhabilis, Br. mammillata, Br. ochotensis, Br. villosa, Buskiella vitjasi, Daylithos parmatus, Flabelligera affinis, Pherusa moorei, Ph. nipponica, Ph. papillata, Piromis suni, Semiodera nishii, and Stylarioides longisetosa (Imajima 1964, 2006, 2009 and Hartman 1964; Uchida 1992; Jirkov 2001; Salazar-Vallejo 2011a, b, 2012, 2014; Miura 2014).

The flabelligerid genus Diplocirrus Haase, 1915 is characterized by having a club-shaped body, two types of retractable branchiae, and multiarticulate chaetae in both parapodia rami. After the comprehensive taxonomic revision by Salazar-Vallejo and Buzhinskaja (2011), the two monotypic genera Bradiella Rullier, 1965 and Diversibranchius Buzhinskaja, 1994 were synonymized with Diplocirrus; consequently, the genus currently consists of 13 species, of which three are still undescribed (Salazar-Vallejo and Buzhinskaja 2011), that live on soft bottoms in sublittoral depths worldwide.

Diplocirrus nicolaji (originally combined with the newly established genus Diversibranchius) was described based on specimens collected from shallow waters of Peter the Great Bay, the Sea of Japan, Russia (Buzhinskaja 1994). Thereafter, an unidentified “Flabelligeridae from Japan” was illustrated in Rouse and Pleijel (2001), which was later identified as D. nicolaji by Salazar-Vallejo and Buzhinskaja (2011) on the basis of their reexamination of the photographed material of Rouse and Pleijel (2001) (Salazar-Vallejo, personal communication). However, neither Rouse and Pleijel (2001) nor Salazar-Vallejo and Buzhinskaja (2011) provided the precise collection sites of their material; no other record of this species has been reported from Japanese waters. Thus, the distribution of D. nicolaji in Japan needs to be clarified.

During the course of a survey of Japanese polychaete fauna by the first and second authors, some flabelligerid worms were collected from four localities of Japan (Oshoro, Tateyama, Misaki, and Komatsubara). After morphological observation, we identified these flabelligerids as D. nicolaji. In this paper, we report the morphology of these Japanese D. nicolaji specimens including observations of the detailed structure of the taxonomic characters using a scanning electron microscope (SEM). In addition, we also obtained partial mitochondrial cytochrome c oxidase subunit I (COXI) sequences for conducting phylogenetic analysis and for contributing to DNA barcoding of Flabelligeridae.

SYSTEMATICS

Family Flabelligeridae de Saint-Joseph, 1894

Genus Diplocirrus Haase, 1915

(New Japanese name: Konbouhabouki-zoku)

Diplocirrus nicolaji (Buzhinskaja, 1994)

(New Japanese name: Bouzu-habouki)

(Figs. 2, 3)

Diversibranchius nicolaji Buzhinskaya, 1994: 231, figures 2–7; Darbyshire and Mackie 2009: 97, Table 1.

Flabelligeridae from Japan: Rouse and Pleijel 2001, plate 11, figure f.

Diplocirrus nicolaji: Salazar-Vallejo and Buzhinskaja 2011, 31–33 figure 9.

Phylogenetic analysis

The final length of the aligned COXI sequence was 531 bp; this contained 230 variable sites, of which 213 sites were parsimony informative. The mean base composition was 28.3 % (A), 20.2 % (C), 16.4 % (G), and 35 % (T).

The ML tree obtained (Fig. 4) showed that the monophyly of Flabelligeridae was strongly supported (bootstrap [BS] value = 95 %); however, the internal relationships within the family were totally unresolved. All the sequences of Diplocirrus nicolaji formed a monophyletic clade with high support value (BS value = 94 %) and were well discriminated from the remains (Fig. 4). The average genetic divergence of the sequences among D. nicolaji was 2.1 ± 1.6 % in p-distance (2.2 ± 2 % in K2P). Monophyly of the two Diplocirrus species included, D. longisetosus and D. nicolaji, was not supported.

The present specimens agreed well with the original and subsequent descriptions of Diplocirrus nicolaji in following diagnostic features: 1) notochaetae in anterior setigers are very short and do not form a cephalic cage; 2) papillae on the tunic are relatively short, digitate or bowling-pin shaped, and arranged in 10–17 rows per segment; 3) ventrolateral gonopores are present from chaetiger 2; 4) neurochaetae in median chaetigers are barely tapered with markedly falcate tips; and 5) the lamellate region of the anterior cirriform branchiae extends to one-third of the branchial length (Buzhinskaja 1994; Salazar-Vallejo and Buzhinskaja 2011). They were thus identified as D. nicolaji with little hesitation. Although a few morphological variations were found among the local populations (see above), our phylogenetic analysis confirmed that the present specimens are conspecific with little genetic divergence (Fig. 4). The differences of palp length among each population might be attributable to the various anesthesia durations. Meanwhile, our analysis based only on COXI sequences did not resolve the inter-family relationships of Flabelligeridae as shown in Osborn and Rouse (2011), which might reflect the fact that the COXI is relatively unconserved. More gene sequences from both the nuclear and mitochondrial genomes, as well as increased taxonomic samplings, will be needed for the better resolution.

Imajima (1992) and Uchida (1992) also reported the genus Diplocirrus from Japanese waters; the former recorded Diplocirrus sp. from shallow waters off Sarufutsu, northern Hokkaido, and the latter merely included this genus in his key of shallow-water flabelligerids in Japan. Unfortunately, we could not ascertain their identification nor compare these records with D. nicolaji since they gave no specific accounts of their species. Consequently, this is the first report of the detailed description of the morphology and distribution of the genus Diplocirrus, as well as D. nicolaji, from Japan.

In this study, D. nicolaji was collected from shallow waters (up to a depth of 10 m) along Japanese coasts (Fig. 1). This is the first report of D. nicolaji from the Pacific Ocean and extends its distribution southward (Komatsubara, Seto Inland Sea). Therefore the distribution of D. nicolaji is not restricted to northern cold waters in the Sea of Japan. Although intertidal to shallow-water polychaete faunae in central to southern Japan have been relatively well surveyed (e.g., Inaba 1988; Nishi et al. 2001; Nishi and Kudo 2003, 2005; Sato 2000; Yamanishi and Sato 2007; Nishi and Kupriyanova; 2011), D. nicolaji was not recorded in those studies and may thus have been overlooked. Future surveys will add its new distributional records in Japanese waters and the vicinity.

This study contributes to the pool of knowledge on Japanese coastal biodiversity and will help generate more accurate environmental assessments. Furthermore, the given COXI sequences will be helpful for the future accurate identification of this species, that is, DNA barcoding. Recently, a number of cryptic species have been discovered through DNA barcoding analyses, even among polychaetes (see Nygren 2014). We have taken a first step toward unraveling the cryptic diversity of Flabelligeridae occurring in Japanese waters.