¾    ¿ÀÁ (  ) - CTOZ

A synoptic review of the family (: Nudibranchia): a multilevel organismal diversity approach

Tatiana Korshunova Koltzov Institute of Developmental Biology, Vavilova str. 26, 119334 Moscow, Russia

Torkild Bakken NTNU University Museum, Norwegian University of Science and Technology, Trondheim, Norway

Viktor V. Grøtan Ørland Froskemannsklubb (Ørland Diveclub), Postboks 197, 7129 Brekstad, Norway

Kjetil B. Johnson Ørland Froskemannsklubb (Ørland Diveclub), Postboks 197, 7129 Brekstad, Norway

Kennet Lundin Gothenburg Natural History Museum, Box 7283, 40235, Gothenburg, Sweden Gothenburg Global Biodiversity Centre, Box 461, 40530, Gothenburg, Sweden

Alexander Martynov Zoological Museum of the Moscow State University, Bolshaya Nikitskaya Str. 6 125009 Moscow, Russia [email protected]

Abstract

A synoptic review of the family Dendronotidae is presented based on morphological and molecular data. Three genera are recognized: , Pseudobornella, and Cabangus gen. nov. Two new Dendronotus are described, D. yrjargul sp. nov. and D. nordenskioeldi sp. nov., which reveal º»ne-scale di¼ferences. Dendronotus yrjargul sp. nov. from mid-Norway and the Arctic regions is a sister species to the North Paciº»c D. kalikal. These two species are showing clear morphological and ontogenetic di¼ferences but are close in genetic distance. In contrast, Dendronotus nordenskioeldi sp. nov. from the Laptev Sea is externally similar to the white morphs of D. lacteus or D. frondosus, but according to the molecular data and radular

©   .,   |  : ./-  This is an open access article distributed under the terms of the 4.0 license.  $        # $# !"%   %   ./-  |   . morphology it is distinct from any of its congenerics. Comparison of molecular and morphological data of D. niveus from the type locality (White Sea) and material from other localities with those from the American North Atlantic coast (type locality of D. elegans) reveals their substantial similarity. Therefore, D. niveus is considered a junior synonym of D. elegans. The present review of the family Dendronotidae contributes to a general discussion on the species concepts and on a recent proposal of multilevel organismal diversity.

Keywords

Cabangus – Dendronotus – Pseudobornella – molecular phylogeny – species problem –

Zoobank: http://urn:lsid:zoobank.org:pub:5E6F2529-E533-4A89-9EE4-5677865C5468

Introduction criterion of reproductive isolation (Mayr, 1969, 1982) is not robust as has been shown The evolutionary process is traditionally con- on numerous occasions (e.g., McDonald and sidered to be continuous (Darwin, 1859) and Koehn, 1988; Mallet, 2005; Kijewski et al., 2006; this concept prevails despite many years of Harrison & Larson, 2014; Jančúchová-Lásková controversial discussions (Eldredge & Gould, et al., 2015; Crossman et al., 2016; Fraïsse et al., 1972; Mayr, 1982; Mallet et al., 2009; Coates et 2016; Potkamp & Fransen, 2019). al., 2018). However, the species itself is sup- With the dawn of the molecular era in posed to be an unavoidably separate unit taxonomy, the problem of hidden diversity according to the nomenclature of current tax- emerged when molecularly divergent groups onomic practice, which in zoology is governed of specimens were discovered within mor- by the ICZN (1999). This is one of the greatest phologically highly similar species as deº»ned contradictions between the evolutionary pro- in the traditional way (Bickford, 2007; cesses and taxonomic classiº»cation, which Korshunova et al., 2017a, 2019a). The discord- has not been solved satisfactorily (Zachos, ant development of morphological structures 2018). In many cases it has been shown that and genetic apparatus per se is also a sign of it is very di¼º»cult to assign a particular spec- evolutionary processes in progress. However, imen to a particular species. Until now there it is still possible to separate such cases by are no unequivocal and universal criteria for using molecular data and if there are exten- species delimitations when the underlying sive genetic di¼ferences, we may assume that evolutionary processes and phylogenies are evolutionary divergence has occurred among taken into consideration (e.g., Zachos et al., such species. Relatively rarerely noticed and 2013; González et al., 2018; Stanton et al., 2019; much more interesting are those cases when Platania et al., 2020), and even for the exist- despite relatively low genetic di¼ferences ence of species at all (Ereshefsky, 1998; Willis, some species exhibit distinct morphological 2017; Reydon & Kunz, 2019). At the conceptual features (Mayer & Helversen, 2001; Wiemers level, more than 26 di¼ferent species concepts & Fiedler, 2007). The current practice of have been suggested (Zachos, 2016) but at a deº»ning potential species by using molecu- practical level even the most widely acclaimed lar markers depending on various distances

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   and factors (e.g., Ratnasingham & Hebert, ontogenetic data in the framework of mul- 2013) referred to as “operational taxonomic tilevel organismal diversity (Korshunova et units” (s) does not consistently corre- al., 2019a). We therefore develop a general spond to taxa at the species level. There is taxonomic framework for phylogeny and an unfortunate discordance in how the tax- taxonomy of the family Dendronotidae, onomic practice currently works because it which for the º»rst time presents a synoptic is based on a classiº»cation system consist- review of all currently known species of the ing of rigid names that does not fully reÐÑect Dendronotus. Its also conº»rms the information of the underlying evolutionary validity of the genus Pseudobornella Baba, processes. A solution for the species prob- 1932 and supports the proposal of the new lem using the separation between metap- genus Cabangus (see details in Synopsis). opulation lineages potentially without any distinguishing characters (de Queiroz, 2006, 2007) contradicts available biological evi- Materials and methods dence, e.g., that even clones show minor but detectable morphological and epigenetic Collecting data di¼ferences. Therefore, to avoid an artiº»cial Specimens were collected by Viktor V. contrast between “cryptic” and “non-cryp- Grøtan and Kjetil B. Johnson in the vicin- tic” species in a classiº»cation and to encom- ities of Trondheim, Norway (63°41’8.66”N pass various available data, the concept 9°39’58.79” E), and also in the Arctic waters of the multilevel organismal diversity was during vessel expeditions, and for compara- proposed (Korshunova et al., 2019a). In the tive purposes by scuba diving in NE Atlantic, present study we apply this approach to the NW and NE Paciº»c. Specimens studied are family Dendronotidae. deposited in the collections at Zoological Currently there is an accelerating dis- Museum, Moscow State University (×ØØ), covery of º»ne-scale species-level diver- and the Norwegian University of Science sity within the genus Dendronotus. In the and Technology, NTNU University Museum last decade, 10 of 12 new Dendronotus spp. (-Ø) (Bakken et al., 2020). were described by our research group using both morphological and molecular data Morphological analysis (Martynov 2015a, 2020a; Ekimova et al. 2015; External and internal morphology was stud- Lundin et al., 2017; Korshunova et al. 2016a, ied under a stereomicroscope using a Nikon b, 2017b, 2019b). In this study, we present a D-810 digital camera and scanning electron modern taxonomic synopsis of the family microscopes. The buccal mass of each spec- Dendronotidae, which together with molec- imen was extracted and processed in 10% ular phylogenetic analyses demonstrates sodium hypochlorite solution to extract the both lower and higher degrees of evolution- radula and the jaws. The jaws of each spe- ary separations as a reliable example of mul- cies were analysed under a stereomicro- tilevel organismal diversity. For instance, scope and then photographed. The jaws and a newly discovered species from Norway radulae were coated, then examined and (Dendronotus yrjargul sp. nov.) already has photographed using a scanning electron developed deº»nite signs of morphological microscope (CamScan, JSM). The reproduc- and molecular divergence. This highlights tive systems were also examined using a the importance of considering broadly stereomicroscope.

 $        # $# !"%   %   ./-  |   .

Molecular analysis (Ronquist et al., 2012). Four Markov chains Eleven specimens were sequenced for the were sampled at intervals of 500 generations. mitochondrial genes cytochrome c oxi- Analysis was started with random starting dase subunit I (Û ) and 16S rRNA, and the trees and 5 × 106 generations. Maximum nuclear genes 28S rRNA (C1-C2 domain). Likelihood-based phylogeny inference was Small pieces of tissue were used for  performed using RAxML 7.2.8 (Stamatakis extraction with Diatom™  Prep 100 kit by et al., 2008)) with 1000 bootstrap replicates. Isogene Lab, according to the producer’s pro- Final phylogenetic tree images were rendered tocols. Standard ÚÛ ampliº»cation options, in FigTree 1.4.2. Nodes in the phylogenetic and sequence obtainment were applied (for trees with Bayesian posterior values > 0.96% details, see Korshunova et al., 2016a, 2017a). and bootstrap values >90% were considered Û sequences were translated into amino ‘highly’ supported. Nodes with 0.90–0.95% acids to verify coding regions and avoid and 80–89% accordingly were considered errors. All new sequences were deposited in ‘moderately’ supported (lower support values GenBank (Appendix, table A1, highlighted were considered not signiº»cant). in bold). Publicly available sequences Û , The program MEGA7 (Kumar et al., 2016) 16S rRNA, 28S rRNA, and additionally the was used to calculate the uncorrected p-dis- nuclear gene Histone 3 (H3) of Dendronotus tances between all the sequences as well as species plus several outgroup taxa (, mean intragroup and intergroup genetic dis- Marionia, Doto, Scyllaea, Crosslandia, Tethys tances. Automatic Barcode Gap Discovery and Notobryon) were also included in the (À ) (Puillandre et al., 2012) (https://bio- molecular analysis. Phylogenetic analysis was info.mnhn.fr/abi/public/abgd/abgdweb.html) performed using 91 specimens of the genus was used to estimate Dendronotus species Dendronotus, three Pseudobornella orientalis divergence. Alignment from the Û marker Baba, 1932 specimens, and 12 outgroup spec- were submitted and processed in À using imens (Appendix, table A1). Sequences were the Jukes-Cantor (JC69) and Kimura (K80) aligned with the MAFFT algorithm (Katoh models with the following settings: a prior et al., 2002). Separate analyses were con- for the maximum value of intraspeciº»c diver- ducted for Û (658 bp), 16S (442 bp), 28S gence between 0.001 and 0.1, 10 recursive (349 bp), H3 (328 bp), and the concatenated steps within the primary partitions deº»ned by dataset (1777 bp). Gblocks 0.91b (Talavera the º»rst estimated gap, and a gap width of 1. and Castresana 2007) was applied to discard A multi-locus species delimitation analysis poorly aligned regions for the 16S data set was performed using BPP v.3.1 (Yang 2015) for (using less stringent options; in total, 5% of Û , 16S, and 28S genes. Two analyses (A and B) the positions were eliminated). The GTR+I+G with di¼ferent population size (θs) and diver- model was chosen for the Û , 16S, and con- gence time (τ0) priors were preformed, using catenated datasets; the GTR+G model was the same settings and priors as in Martinsson chosen for the 28S and H3 datasets using & Erséus (2018) (X: θ 2,400, τ0 2,200; Y: θ MrModelTest 2.3 (Nylander et al., 2004). 2,1000, τ0 2,200). All analyses were performed Two di¼ferent phylogenetic methods, three times to conº»rm consistency between Bayesian Inference ( ) and Maximum runs. We considered species delimited with a Likelihood (Ø), were used to infer evolution- ÚÚ ≥ 0.90 in all analyses to be well supported. ary relationships. Bayesian estimation of poste- For clusters with a ÚÚ < 0.90, we accepted the rior probability was performed in MrBayes 3.2 best-supported more inclusive species.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Results placed P. orientalis (ÚÚ = 0.99) as sister to the genus Dendronotus. The H3 single tree shows Molecular phylogenetic relationships polyphyly in genus Dendronotus if Cabangus amongst species of the genus Dendronotus regius would be classiº»ed as Dendronotus. For Bayesian Inference ( ) and Maximum P. orientalis, the minimal mean Û intergroup Likelihood (Ø) analyses based on the com- genetic distance was found between P. orienta- bined dataset for the mitochondrial Û lis and D. velifer (15.6%), and maximal between and 16S, and the nuclear 28S and H3 genes P. orientalis and D. lacteus (18.3%). For C. yielded similar results (º»g. 1). The family regius, the minimal mean intergroup genetic Dendronotidae formed a highly supported distance was found between C. regius and D. clade with  and Ø ( = 100, ÚÚ = 0.96). kalikal (16.2%), and the maximal between Dendronotus elegans Verrill, 1880 from C. regius and D. frondosus (18.4%). The mean Newfoundland and Greenland (close to the intergroup genetic distance within the clade type locality of this species) nested together including D. robustus, D. patricki and D. velifer with D. elegans (= D. niveus Ekimova et al., ranged from 7.2 to 16.8%. The mean inter- 2015 syn. nov.) from the White, Barents, Kara, group genetic distance within the clade “other and Laptev seas in a maximum-supported Dendronotus” ranged between 2.1 and 14.2% separate clade (ÚÚ = 1,  = 100). Dendronotus (Appendix, table A2). nordenskioeldi sp. nov. shows a sister posi- To deº»ne species, we used an integra- tion with the D. europaeus Korshunova et al., tive approach, including phylogenetic tree 2017 clade and the clade including D. rufus topologies, À and ÚÚ analyses, pairwise O’Donoghue, 1921 and D. lacteus (Thompson, distances as well as the haplotype network 1840), but without good node support. for the molecular data reconstructed using Evolutionary ties for D. nordenskioeldi sp. nov. Population Analysis with Reticulate Trees are not fully clariº»ed. Dendronotus kalikal (PopART). Ekimova et al., 2015 and D. yrjargul sp. nov. To delimit the species based on single locus clustered in separate clades. Node support sequence data (Û ), species delimitation was for the D. yrjargul sp. nov. clade is moderate performed with the À tool. In the full data (ÚÚ = 0.93,  = 78). Support for the D. kalikal set (88 Dendronotus specimens), results from clade is highly supported with  and moder- initial partitions show 20 potential species: 16 ately supported with Ø (ÚÚ = 0.99,  = 83). reputed Dendronotus species + D. elegans from Individual gene trees showed similar the type geographic region (North Atlantic), trends, however the position of Pseudobornella combined together with D. elegans (“as D. orientalis and Cabangus regius (Pola & Stout, niveus”) from the Arctic region + D. rufus, 2008). varied (º»g. 2). The Û single tree combined with D. lacteus + D. nordenskioeldi placed C. regius (ÚÚ = 0.99) as sister to the sp. nov. + D. yrjargul sp. nov., combined with genus Dendronotus, and P. orientalis as sister D. kalikal. Generally, recursive partitions have to these together. The 16S single tree showed more groups than initial partitions that are that P. orientalis forms a weakly supported typically stable. Results from recursive parti- polytomy (ÚÚ = 0.6) with the clade contain- tions show 22 potential species, correspond- ing D. robustus, D. patricki Stout et al., 2011 ing the potential Dendronotus species clades and D. velifer, and the clade including other on the phylogenetic tree (º»g. 1): 16 reputed Dendronotus species. C. regius was sister to all Dendronotus species in additon to separate these clades (ÚÚ = 0.98). The 28S single tree clades for D. nordenskioeldi sp. nov., D. yrjargul

 $        # $# !"%   %   ./-  |   .

ë À 1 Phylogenetic relationships of the species of the genus Dendronotus based on Û + 16S + 28S + H3 concatenated dataset inferred by Bayesian Inference ( ). Numbers represent posterior probabilities from  and bootstrap values for Maximum Likelihood. Some branches are collapsed at species level. Dendronotus elegans from the type region (North Atlantic) are highlighted in dark purple. Abbreviations: h – holotype; p – paratype. Summary of species delimitation results from À analyses for the Û dataset is shown.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

ë À 2 Phylogenetic relationships of the species of the genera Dendronotus, Pseudobornella and Cabangus in single Û , 16S, 28S, and H3 gene trees inferred by Bayesian Inference ( ). Some branches are collapsed. Positions the Pseudobornella orientalis highlighted green; positions the Cabangus regius highlighted orange. Numbers represent posterior probabilities from  . ÚÀÚ Á . ØÁ  . 

 $        # $# !"%   %   ./-  |   . sp. nov., D. kalikal, D. rufus, D. lacteus, and Systematic account D. elegans from the North Atlantic, together Descriptions of the new species of the genus with D. elegans (as “D. niveus”) from the Arctic Dendronotus region. Analyzing multi-locus genomic sequence Dendronotus yrjargul sp. nov. data under the multispecies coalescent model Figs 1–4, 6, 7 was conducted for D. elegans (2 specimens), ZooBank: http://urn:lsid:zoobank.org:act: former D. “niveus” (16 specimens), D. yrjargul D44EDB5D- 9C27-4D38-A18B-AD6DA5ACE43D sp. nov (11 specimens). and D. kalikal (10 spec- imens). In analysis A, the three species model Etymology. From Norwegian yrjar (= old (D. elegans combined with the former “D. name for the type locality in Ørland) and gul niveus”, separate D. yrjargul sp. nov. and sep- (= yellow) meaning “yellow/golden of Ørland” arate D. kalikal) is preferred with a mean ÚÚ in reference to the remarkable habitus of this of 0.94. In analysis B, the same three species species. model is also preferred with a higher mean ÚÚ of 0.99. Type locality. Norwegian Sea, Norway, Ørland. Results obtained by PopART for the Û and 16S genes showed a network of haplo- Material examined. Holotype. NTNU-VM- types that clearly clustered into three groups 76306, live length 95 mm (48 mm preserved), coincident with D. elegans combined with Norwegian Sea, mouth of Trondheims ºìord, former “D. niveus”, D. yrjargul sp. nov. and D. Ørland, Breivika, 12.03.2018, depth 20–25 m, kalikal (º»g. 3). collector Viktor V. Grøtan. Paratypes with Regarding the supposedly fast-evolving same provenance data: NTNU-VM-76307, Û marker, genetic distance values between 1 specimen, live length 75 mm (40 mm D. elegans from Newfoundland and D. elegans preserved); NTNU-VM-76308, 1 specimen, live from Greenland is 0.61%. Maximal intragroup length 60 mm (40 mm preserved); paratypes distance among specimens formerly iden- with same provenance data but collected tiº»ed as “D. niveus” from the White, Barents, 08.01.2018, depth 20–25 m, collector Viktor V. Kara, and Laptev seas is 1.42%, whereas dis- Grøtan, Kjetil B. Johnson: NTNU-VM-76302, 1 tances between the D. elegans and specimens specimen, preserved length 18 mm; NTNU-VM- formerly identiº»ed as “D. niveus” range from 0 76305, 1 specimen, preserved length 17 mm; to 0.94%, indicating without doubt that D. ele- NTNU-VM-76303, 1 specimen, preserved length gans and former “D. niveus” are the same spe- 14 mm; NTNU-VM-76304, 1 specimen, preserved cies. Mean intragroup distance within the D. length 19 mm; ZMMU Op-718, 1 specimen, elegans clade is 0.59% whereas minimal inter- preserved length 19 mm. group distance (7.6%) was found between D. elegans and D. dalli (Appendix, table A2). Description. Body elongate, living specimens The molecular phylogenetic results con- up to 95 mm in length (º»g. 4). Five to six º»rm the morphological analysis data (see main branched appendages on oral veil. Five Systematic section and º»gs 1–8). Thus, D. appendages on rhinophoral stalks; 18–30 elegans and “D. niveus” are the same species, rhinophoral lamellae. Branched lateral papilla on whereas D. kalikal, D. yrjargul sp. nov, and D. rhinophoral sheath present. Seven to eight pairs nordenskioeldi sp. nov. are three separate spe- of dorsolateral appendages present in large adult cies in the genus Dendronotus. specimens. Lip papillae 10 to 20. Dorsolateral

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

ë À 3 A. Map of the Northern Hemisphere showing the distribution of D. elegans (including “D. niveus”), D. kalikal, and D. yrjargul sp. nov. The type locality of D. elegans is indicated by crossed lines. B. The haplotype networks based on cytochrome c oxidase subunit I (Û ), 16S rRNA (16S), and 28S rRNA (28S) molecular data showing genetic mutations occurring within D. elegans (“D. niveus”), D. kalikal, and D. yrjargul sp. nov. specimens. Abbreviations: BeSt – Bering Strait;  – Barents Sea; Gr – Greenland; Km – Kamchatka;  – Kara Sea; Ku – Kuril Islands;  – Laptev Sea; Nf – Newfoundland;  – Norwegian Sea; í – White Sea. Type locality of D. elegans is marked on the map by “x”.

 $        # $# !"%   %   ./-  |   .

ë À 4 Dendronotus yrjargul sp. nov. (a–n) and D. kalikal (o–t), external and internal morphology. a–c: living holotype of D. yrjargul NTNU-VM-76306 (a) and its habitat (b, c), dorsal, ventral and lateral views with egg mass and associated hydroid Abietinaria fusca, respectively; d: Dendronotus yrjargul, habitat of living paratype NTNU-VM-76302; e: jaw of holotype; f: denticles on the masticatory edge of the holotype; g: jaws of paratype NTNU-VM-76302; h: discoid ridge-like structures on the masticatory edge of paratype NTNU-VM-76302; i: details of discoid structures on the masticatory edge of paratype NTNU-VM-76302; j: anterior part of radula of holotype; k: central teeth of radula of holotype; l: lateral teeth of radula of holotype; m: central teeth of radula of paratype NTNU-VM-76302; n: lateral teeth of radula of holotype; o: living holotype of D. kalikal; p: jaw of D. kalikal, specimen from Kurile Islands;

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   appendages with relatively short primary stalk, We have only rarely observed the new species strongly branched secondary branches and on other hydroids. Dendronotus yrjargul sp. nov. elongate tertiary branches (º»g. 4A–D). Basal appears quite abundant in this location, up to colour is semitransclucent white to greyish, 10 individuals per m2. The new species appears with characteristic two thin wavy subparallel to have a one-year life cycle. Mating and egg brownish-orange to brownish lines running laying occurs around March, and the large adult from the head nearly to the tail. There are no individuals eventually disappear throughout specimens with brownish basal colour. Upper the spring, as does the hydroid they live on. part of the dorsolateral appendages covered During the summer, small individuals begin to with bright golden yellow to yellowish pigment. reappear, along with the new growing phase of Foot bordered with thin brownish-orange line. A. fusca. The new generation of D. yrjargul sp. Dorsal processes of jaws inclined posteriorly nov. grow throughout autumn and winter. at approximately 40–48° to the longitudinal axis of the jaw body and 0.4 of its length (º»g. Remarks. Externally, D. yrjargul sp. nov. readily 4E–I). Masticatory border of jaw with evident di¼fers from all known species of the genus ridge-like denticles (º»g. 4H, I). Radular formula: Dendronotus (synopsis given below), including 38–50 × 0–9.1.9–0. Central (rachidian) tooth the sister species from the North Paciº»c, D. kalikal moderately denticulated and bearing up to 16 (description for comparison is given below), by a distinct denticles (º»g. 4J–M). Denticles with deep combination of o¼f-white semitransparent body furrows. Lateral teeth elongate, slightly curved, with two thin dorsal subparallel brownish lines bearing up to 7 denticles (º»g. 4J–N). Ampulla and dorso-lateral appendages golden-yellow to thickened kidney-shaped (º»g. 6A). Prostate yellowish (º»g. 4A–D). Instead, D. kalikal invariably discoidal, consisting of up to 50 alveolar glands. possesses a basal body colour covered with Distal part of vas deferens moderate in length, numerous brownish speckles and small spots, transitioning to long, thick, curved copulatory which make subparallel dorsal lines considerably organ. Bursa copulatrix large, rounded, with blurred and indistinct (º»g. 4O). Importantly, we seminal receptaculum placed distally. investigated more than ten D. kalikal specimens from very distantly placed localities in the Distribution. From Norwegian Sea to Kara Sea. Bering Strait, Kamchatka and Kurile Islands (º»g. 3A) and not one specimen possessed a mostly Habitat and ecological observations. Dendronotus whitish semitransparent body with distinct thin yrjargul sp. nov. is closely associated with the subparallel dorsal lines, as in D. yrjargul sp. nov. hydroid Abietinaria fusca (Johnston, 1847) (see (º»g. 4A–D). The whitish colouration in D. kalikal Vervoort, 1942) which forms dark brown colonies consists of a background colour represented by that resemble small dry º»r branches (personal areas in between the numerous basal brownish observations of VG and KJ). The hydroid usually spots and stripes. Furthermore, considerable inhabits somewhat deeper water, but currents di¼ferences in the reproductive system and the and other conditions in the Norwegian waters maturation process between D. yrjargul sp. nov. make it reachable by scuba diving at circa 20 m. and D. kalikal were revealed in course of the

ë À 4 q: details of the masticatory edge of the same specimen; r: anterior part of radula of the same specimen; (cont.) s: central teeth of radula, same specimen; t: lateral teeth of radula, same specimen. Scale bars: e: 300 µm; f: 100 µm; g: 500 µm; h, m, n, q, t: 20 µm; i: 5 µm; j, k, r, s: 50 µm; l, t: 20 µm; p: 200 µm. ÚÀÚ Á . Àï (– )  .  (). Ø Ø ÛÀÚ Á . ØÁ

 $        # $# !"%   %   ./-  |   . present study. All studied specimens of D. kalikal had been originally named “Nordenskiöld have an elongated bursa copulatrix (º»g. 6B), Sea”, after this Arctic explorer. whereas in all studied specimens of D. yrjargul sp. nov. (º»g. 6A) the bursa copulatrix is rounded. Type locality. Laptev Sea. Furthermore, we discovered that there is considerable di¼ference in the patterns of Type material. Holotype. ZMMU Op-665. maturation (and hence, ontogenetic patterns) Arctic region, Laptev Sea, sta. 100, 69 m depth, between D. yrjargul sp. nov. and D. kalikal. collector Olga Zimina. Dendronotus yrjargul sp. nov. reaches a big size of 95 mm (live) in the adult state, and in Description. Body relatively narrow, preserved specimens of 20–30 mm (live) the reproduc- holotype 19 mm length (º»g. 5). Five to six tive system was still very weakly developed. main branched appendages on oral veil. Instead, D. kalikal has a maximum length of Five appendages on rhinophoral stalks; 17 mm when preserved and already at lengths 12–14 rhinophoral lamellae. Lateral papilla of 7–10 mm live specimens show a well-de- on rhinophoral sheath present. Five pairs of veloped reproductive system. The molecular dorsolateral appendages. Circa 20 lip papillae. di¼ferences between D. yrjargul sp. nov. and Dorsolateral appendages with relatively D. kalikal are as follows. Regarding the Û short primary stalk, moderately branched marker, the maximal intragroup distance secondary branches and tertiary branches (º»g. within D. kalikal is 1.74% and within D. yrjargul 5A, B). Basal colour is semitranslucent white. sp. nov. it is 0.85% (mean intragroup distances No additional pigment. Dorsal processes of are 0.80% and 0.39%). The mean intergroup jaws inclined posteriorly at approximately 53° distance between D. kalikal and D. yrjargul to the longitudinal axis of the jaw body and sp. nov. groups is 2.1% (Appendix, table A2). 0.42 of its length (º»g. 5D). Masticatory border Thus, intragroup distances within D. yirargul of jaw with º»ne denticles (º»g. 5E). Radular sp. nov. are smaller than the distance between formula: 43 × 3–11.1.11–3. Central (rachidian) D. yirargul sp. nov. and D. kalikal. Thus, D. tooth bears very weak denticulation that are yrjargul sp. nov. and D. kalikal demonstrate di¼º»cult to separate into individual denticles considerable di¼ferences in morphological, (up to 20 and more variously varied denticles molecular and, to a very large degree, ontoge- at least) (º»g. 5F, G). Denticles without deep netic levels. Dendronotus yrjargul sp. nov. thus furrows. Lateral teeth elongate, slightly curved, represents a well-separated species, although bearing up to seven denticles (+ 2–3 reduced possibly recently diverged, as indicated by the denticles may present) (º»g. 4F–H). Ampulla slight molecular divergence. thickened, kidney-shaped. Prostate discoidal, consisting of up to 20 alveolar glands. Distal Dendronotus nordenskioeldi sp. nov. part of vas deferens moderate in length, Figs 1–3, 5–7 transitioning to long, thick, curved copulatory ZooBank: http://urn:lsid:zoobank.org:act: organ. Bursa copulatrix large, distinctly 9FB63088-BF8A-4990-B292-7E26DEA0FDCF elongated, with seminal receptaculum placed distally. Etymology. In honour of Baron Nils Adolf Erik Nordenskiöld, outstanding Arctic explorer, Distribution. So far known only from the geologist, and mineralogist. The Laptev Sea Laptev Sea.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

ë À 5 Dendronotus nordenskioeldi sp. nov. ZMMU Op-665, external and internal morphology. a–b: living holotype, dorsal and lateral views respectively; c: ventral view of preserved holotype; d: jaw of holotype; e: irregular denticles on the masticatory edge of the holotype; f: posterior part of radula; g: central teeth of radula; h: lateral teeth of radula of holotype. Scale bars: d: 1mm; e: 20 µm; f: 100 µm; g, h: 30 µm. ÚÀÚ Á . × Ø  (, )  .  (Û). Ø Ø ÛÀÚ Á . ØÁ

Remarks. Externally, Dendronotus norden- reticulation and peculiar ribs on the central skioeldi sp. nov. resembles white forms of D. teeth, which is a common characteristic of D. lacteus but di¼fers by the absence of a distinct lacteus (Korshunova et al., 2017a). According

 $        # $# !"%   %   ./-  |   .

ë À 6 Reproductive systems. a, Dendronotus yrjargul sp. nov.; b, D. kalikal; c, D. elegans; d, D. nordenskioeldi sp. nov. Scale bars: a–d 1mm. Abbreviations: am, ampulla; bc, bursa copulatrix; fgm, female gland mass; fo, female opening; p, penis; pr, prostate; rs, receptaculum seminis; ud, uterine duct; vd, vas deferens; vg, vaginal duct. to the molecular di¼ferences, a minimal Synopsis of the previously described valid intergroup distance for the Û gene (7.9%) taxa of the family Dendronotidae is found between D. nordenskioeldi sp. nov and D. europaeus. The mean intergroup Genus Dendronotus Alder & Hancock, 1845 distances between D. nordenskioeldi sp. nov. Type species. and D. rufus, D. lacteus, and D. kamchaticus are (Ascanius, 1774) 8.3%, 8.3%, and 8.8% respectively. A maximal intergroup distance (15.0%) is found between Diagnosis. Dorsolateral appendages with distinct D. nordenskioeldi sp. nov and D. patricki tertiary branches. Oral veil with branched (Appendix, table A2). appendages. Jaws commonly with set of tightly

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

ë À 7 Representaives of the genus Dendronotus (living specimens). ÚÀÚ Á . , . ØÁ, . ëÛ, . Ø , Á. ë í, . Àí, . ØÁ, . ØÁ,  . × Ø 

 $        # $# !"%   %   ./-  |   . packed ridge-like semi-discoid structures and species, an extended diagnosis is provided also with denticles. Radula with at least more when possible. The º»ne-scale diagnostics of than eight lateral teeth in adult specimens. the genus Dendronotus is a complex task since Central teeth with cusp integrated within lateral variation patterns of habitus and anatomical denticles (or within teeth body if denticles characters overlap in various species. By a absent), the central cusp is not protruding. careful listing it is possible to step-by-step Denticle-bear part of central teeth separated fulº»ll the task and present a mutilelevel from lateral sides by usually distinct shoulder, organismal diversity, though not in a strictly thus central teeth with squarish base and deº»ned form, as was targeted previously triangular top. Prostate with evident, conspicous (e.g., Odhner, 1936; Robilliard, 1970), but in alveolar glands, commonly represent by a a complex form that is best suited to the discoid structure (varies from a ÐÑattened disk evolutionary mosaic in character evolution. to a compressed spherical structure), or very There are also still uncertanties regarding rarely non-discoid, linear (in this case alveolar some particular characters as well as previous glands also present). Copulative organ basically overinterpretations. First of all, previously conical, with various ratios of length and width the pattern of branching in the dorsolateral and degrees of apical tapering. appendages was claimed to be of considerable taxonomic importance (Ekimova et al., Natural history. Representaives of the genus 2015: Fig. 8). In this respect, it is necessary to Dendronotus inhabit a broad bathymetric state, that while some trends exist in a few range, from intertidal to depths over 3000 m, species, importance of these characters were and a variety of substrates (including stony considerably overestimated partly due to the and soft benthic environments). A majority use of preserved specimens in the analysis. of the Dendronotus species feed on hydroids This can be illustrated by the D. kamchaticus (Robilliard, 1970; Thompson & Brown, 1984; case, when according to Ekimova et al. (2015: Martynov & Korshunova, 2011). There as a few Fig. 8D), D. kamchaticus has bulbous short notable exceptions, such as D. iris Cooper, 1863 branches that di¼fer considerably from those from the northeastern Paciº»c, which is an active in any other Dendronotus species. This resulted predator of ceriantharian cnidarians (Wobber, from the use of preserved specimens with 1970; Shaw, 1991), whereas for the wide-bodied considerably contracted branches, whereas species D. robustus Verrill, 1870, D. velifer G.O. the living adult specimens of D. kamchaticus Sars, 1878 and D. bathyvela Martynov et al., have elongated branches of dorsolateral 2020a less speciº»c prey spectrum (including appendages, which are not fundamentally polychaetes and hydroids) has been recorded di¼ferent from the type species of the genus, (Roginskaya, 1990, 1997). D. frondosus (Korshunova et al., 2016a). The description of D. zakuro (similar to D. Remarks. The genus Dendronotus is restricted kamchaticus by radula morphology) is based here since some species have been moved to on recently collected specimens with both other genera. See discussions and remarks shorter “bulbous” (see Martynov et al., 2020a: under the genera Cabangus gen. nov. and Fig. 3A) and elongate dorsolateral appendages Pseudobornella. (Martynov et al., 2020a: Fig. 3D, E), which is due to di¼ferences in physiological conditions. The Species composition. Diagnoses and remarks for preserved specimen of D. zakuro (see Martynov 27 included species are listed below. For each et al., 2020a: Fig. 3C) demonstrates very similar

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

“bulbous” branches, which were incorrectly the genus Pseudobornella are mentioned after depicted as speciº»c for D. kamchaticus in its the diagnoses of the species of the genus original description (Ekimova et al., 2015: Dendronotus. Fig. 8D) (see details in Synopsis under D. There are also various uncertainties and kamchaticus). Dendronotus lacteus apparently misinterpretations regarding internal charac- has shorter tertiary branches, and a “curly” ters that need to be addressed before diagno- appearance of the dorsolateral appendages. ses can be provided. For example, Robilliard However, variation in D. lacteus does not only (1970) suggested to apply several coe¼º»cients concern ontogenetic stages, but also occurs of jaw proportions. However, it is easy to among di¼ferent adult specimens. So, some incorrectly apply respective angle of measure- large specimens of D. lacteus have quite long ments and receive very di¼ferent results, as for elongate branches of dorsolateral appendages, example in Ekimova et al. (2015) compared similar to those in D. frondosus (Korshunova to the same species which jaws by Robilliard et al., 2017a). Most species of the genus (1970). As an example, for D. dalli in Ekimova Dendronotus possess considerably branched et al. (2015: 863) an angle of dorsal processes dorsolateral appendages with distinct tertiary to the longitudinal axis of 25° was indicated, branches (MacFarland, 1966; Robilliard, whereas in Robilliard (1970: 454) this was 60°. 1970; Korshunova et al., 2016a, b, 2017a, In the present study we re-checked jaw pro- 2019b; Martynov et al., 2020a). Unbranched portions and presented updated values for the dorsolateral appendages were recently reported majority of species with su¼º»cient available only for D. claugei from the deep-sea (Valdes et information. This character still needs further al., 2018), but due to the most basal position testing, but some characteristic patterns for of this species (º»g. 1) the generic position of di¼ferent species are likely to occur. Similarly, D. claugei needs to be evaluated with more the number of the prostatic alveoles once data. For another deep-sea species, D. comteti, was thought to be a character that well char- the unbranched dorsolateral appendages acterizes di¼ferent species (Robilliard, 1970; were also indicated (Valdes et al., 2018: 413), Ekimova et al., 2016). However, our previous however, according to the original description (Korshunova et al. 2016a, b) and present study of D. comteti, it possesses both branched and show that this character is much more varia- unbranched dorsolateral appendages (Valdes & ble, and while there are clear trends in some Bouchet, 1998: 320). Unbranched dorsolateral species, it should generally be used with care. appendages can occur in posterior parts of some In the species diagnoses below, di¼ferences other Dendronotus species (e.g., in D. robilliardi, in shape, length and proportions of the cop- see Korshunova et al., 2016a). For some species, ulative organ (penis) are indicated. While in e.g., for D. dalli in Ekimova et al. (2015: 862), some species the di¼ferences in penial mor- the indication of absent tertiary branches phology are clear, in other species this can be of dorsolateral appendages is incorrect. partly due to di¼ferent degrees of contraction deº»nitely has tertiary after preservation. This character needs fur- branches (Robilliard, 1970: 453; present study, ther investigation. º»g. 7). Instead, in the genus Pseudobornella the In the present study we show that the shape general appearance of dorsolateral appendages and size of the bursa copulatrix, at least among does not show as branches, but it is ctenidium- several Dendronotus species (e.g., between like, because of very indistinct (or truly absent) D. kalikal and D. yrjargul), are rather con- tertiary branches (º»g. 2). More details about stant and can be used as a reliable diagnostic

 $        # $# !"%   %   ./-  |   . character. However, the exact position of the Martynov et al., 2020a), in the diagnoses commonly very small receptaculum seminis below we do not provide the number because can be very di¼º»cult to determine, but despite it can be misleading, because only a part of that, it will be given further attention in the the row was counted in various species. True diagnoses below. rodlets can occur in addition, as part of the Di¼ferent terms were applied for the struc- thin labial cuticle, which is separated from tures on the masticatory edge within di¼fer- the masticatory edges. Such true rodlets have ent species of the genus Dendronotus. For been well documented for the tropical repre- example, MacFarland (1966) used “denticles” sentatives of the family Dendronotidae (Pola or “ridge-like” denticles and Robilliard (1970) & Stout, 2008), and possibly also occur in applied “denticles”, whereas Ekimova et al. many Dendronotus species, since the labial (2015) used “rodlets”. In this respect, the term cuticle with rodlets were already described for “rodlets” was used for rod-like structures that several representatives (see e.g., MacFarland, constitute the body of jaws in other nudi- 1966). However, these characters remain branch groups. In the genus Dendronotus, the largely untested by modern methods and we elements on the masticatory edges are usually do not include it in the diagnoses below. As termed as “denticles” when the bodies of jaws an additional complication, denticles and are entire. However, in the genus Dendronotus especially ridge-like structures on the masti- there is a special set of structures in addition catory edges can be easily discarded during to the denticles that appears as broad semi- preparation, and therefore in some previous disks or ridges, which can additionally be descriptions it was inappropriately indicated covered with smaller spine- or denticle-like that masticatory edges are devoid of any structures. It is likely that most species of the denticles or ridges. For example, the absence genus Dendronotus possess such structures. In of the denticles on the masticatory edges addition to the semi-disks/ridges at the masti- of D. frondosus were incorrectly mentioned catory edges, more normally looking denticles in Ekimova et al. (2015), which was further can occur. Transitive denticles on masticatory used as a putative di¼ference between D. fron- edges were also occasionally reported for juve- dosus and D. primorjensis (Ekimova et al., niles of D. frondosus s.l. (Colgan, 1914; Wägele 2016). However, in both species denticles on & Willan, 2000). Therefore, in the present the masticatory edges are deº»nitely present study we use two di¼ferent terms “ridge-like (Korshunova et al., 2016a). The same is true structures” and “denticles” when character- also for other species, such as D. kamchaticus izing the masticatory edges of Dendronotus and D. kalikal. For D. kamchaticus the absence species. The few denticles seen in many cases of any structures on the mascticatory edges are likely a precursor that further transits into was indicated in the original description. For semi-disk structures on the masicatory edges D. kalikal the presence of merely “rodlets” was (see e.g., this clearly for D. kalikal, º»g. 4Q). reported in the original description (Ekimova Because the full row of denticles and/or ridges et al., 2015). However in reality denticles and is not documented for all species due to di¼º»- ridge-like structures are present in both of culties during preparation, the exact number these species (Korshunova et al., 2016a, 2019b; for these structures can be uncertain. Though for D. kalikal see º»g. 4Q). Therefore, when the whenever possible, we already presented the original descriptions denticles or ridge-like number of denticles/ridges, for example the structures was not indicated for species that recently described Dendronotus species (e.g., we could not check, we suppose the presence

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   of at least one of these structures and indicate remarkable transformation during the devel- that in the diagnoses. opment from early juvenile to adult. The cen- Though a pair of salivary glands indeed tral tooth in earlier juveniles is low and bears inserts into the pharynx on the sides of the few distinct lateral denticles. The lateral teeth oesophagus, the knowledge of these is limited in early juveniles are very few in number only to a few species (e.g., MacFarland, 1966; (1–3) or absent, showing an indistinct dentic- Valdes & Bouchet, 1998; Pola & Stout, 2008). ulation or they are almost smooth. In individ- The salivary glands are usually not º»gured uals approaching the adults stage, the central in the original descriptions (an exception: teeth become elongate-triangular with dis- Valdes and Bouchet, 1998) and were never tinct shoulders, lateral denticles become consistently used for taxonomic diagnostics. relatively smaller and numerous or com- We therefore do not include salivary glands pletely reduced. Meanwhile, the lateral teeth in the diagnoses of the species of the genus increase in number and the denticles become Dendronotus. more conspicuous. Since parts of juvenile The radular patterns within the genus radula may persist in some adults, this may Dendronotus still o¼fer one of the best diag- create a considerable confusion regarding the nostic characters (Korshunova et al., 2017a). taxonomic importance of radular characters. However in several species groups quite For instance, D. dalli usually has a smooth similar patterns of the central tooth can be central teeth without denticles in adult stage, observed. These similarities occur between whereas for adults of D. dalli also denticles on groups/clades of closely related species (for central teeth were reported, which led to the example between species that closely related incorrect synonymyzation of D. dalli with D. to D. frondosus, e.g., D. primorjensis and D. frondosus (Odhner, 1936; MacFarland, 1966). venustus, between D. lacteus and D. rufus, This decision was later changed in favour of between D. dalli and D. elegans), but also the validity of D. dalli, but uncertanties over among apparently more distantly related spe- the taxonomic application of the putatively cies such as D. kamchaticus and D. zakuro, and variable radula remained (Robilliard, 1970; between D. albus and D. robilliardi (see º»g. 1). Roginskaya, 1987). Such “variations” can Despite these similarities, in a majority of be overcome by a consistent application these cases º»ne-scale di¼ferences in morphol- of general ontogenetic principles in taxon- ogy of radula can be detected. At the same omy, using the attribution of a particular time, di¼ferent species which are grouped in set of characters to a particular ontogenetic the same subclade (º»g. 1) can demonstrate stage (Martynov et al., 2015b, 2020b). Thus, considerably di¼ferent radular patterns, as for the application of a multilevel diagnostic example the clade consisting of D. jamsteci, within the genus Dendronotus (one that D. subramosus and D. albus (Robilliard, 1970; combines morphological characters of vari- Pola & Stout, 2008; Korshunova et al., 2016a; ous levels and molecular data) instead of a Martynov et al., 2020a). Furthermore, the rad- search for immediate and clearly distinctive ulae of the morphologically very di¼ferent D. di¼ferences, requires a º»ne-scale, complex robilliardi and D. gracilis demonstrate similar approach. patterns in the presence of a widened central tooth with numerous small lateral denticles Dendronotus albopunctatus Robilliard, 1972 without furrows (Baba, 1949; Korshunova Dendronotus albopunctatus Robilliard, 1972: et al. 2016a, 2019b). The radula undergoes 421–430, Figs 1–6; Behrens, 1980: 72; Behrens

 $        # $# !"%   %   ./-  |   . and Hermosillo, 2005: 93; Lamb and Hanby, MacFarland, 1966 2005: 264. Fig. 7 Dendronotus albus MacFarland, 1966: 275– Extended diagnosis. Body wide. Four to º»ve 279, Pl. 40, Fig. 1; Pl. 46, Figs 1–4; Pl. 47, Figs (plus one–two posterior) pairs of dorsolateral 8–11; Pl. 48, Figs 7, 8; Pl. 49, Fig, 5; Pl. 50, Fig. 4; appendages. Six to 12 appendages of oral Pl. 51, Figs 6, 7. veil. Three to seven appendages (equal in Dendronotus diversicolor Robilliard, 1970: size) of rhinophoral stalks. Lateral papilla of 470–474, Figs 4–6, 25–28, Pl. 64, Figs 35, 36. rhinophoral sheaths absent. Rhinophores Not D. albus sensu Robilliard, 1970: 466–470; with 9–17 lamellae. No lip papillae reported Behrens, 1980: 72; Behrens and Hermosillo, in the º»rst description. Basal colour yellow- 2005: 93; McDonald, 1983: 172; McDonald, white, pale pink to red-brown, with red-brown 2009: 463 = D. robilliardi Korshunova, spots and tiny opaque white spots scattered Sanamyan, Zimina, Fletcher, Martynov, 2016. over whole body, solid white line along edge of foot and oral veil. Dorsal processes of Extended diagnosis. Body relatively narrow. jaws inclined posteriorly at approximately Four to º»ve pairs of branched dorsolateral 50° to the longitudinal axis of the jaw body. appendages, digestive gland penetrates only Masticatory processes bear distinct denticles two pairs of dorsolateral appendages. Up to which increase in size towards the distal end six appendages of oral veil. Three to seven of the process. Radula with up to 41 rows of appendages (external longest) of rhinophoral teeth. Central tooth with up to 25 denticles. stalks. Lateral papilla of rhinophoral sheaths Up to 14 lateral teeth with up to 12 denticles. present. Rhinophores with 12–21 lamellae. Ampulla voluminous, folded. Bursa copulatrix Two to eight lip papillae. Basal colour large, oval. Seminal receptaculum very small translucent white to violet, dorsolateral and placed distally at a short distance from appendage variable opaque orange or opaque the vaginal opening. Prostate discoid with white, including tip. Dorsal processes of about 11–16 alveolar glands. The vas deferens jaws inclined posteriorly at approximately is long, penis short, conical. Body length up to 50–60° to the longitudinal axis of the jaw 60 mm. body and 0.4–0.47 of its length. Masticatory borders apparently with ridge-like structures Distribution. NE Paciº»c, Washington and and denticles. Radula with up to 38 rows Oregon, USA. of teeth. Central tooth with up to 25 small distinct denticles without furrows. Up to nine Bathymetry. Depths of 18–40 m (soft lateral teeth with up to 14 denticles. Ampulla subs tra tum). voluminous, folded. Bursa copulatrix large, oval. Seminal receptaculum very small and Remarks. Externally, D. albopunctatus is placed distally at a short distance from the somewhat similar to other species that also vaginal opening. Prostate discoid with about inhabit soft substrata, such as D. robustus 10–30 alveolar glands. The vas deferens is and D. velifer. However, its radular characters relatively short, bent, penis short, nearly are more similar to those of the majority of straight, tapering to a blunt tip. Body length Dendronotus species. up to 73 mm.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Distribution. NE Paciº»c: California to southern large and rounded. Seminal receptaculum Alaska, Canada and US. Records from Baja small placed distally at a moderately short California, Mexico, need veriº»cation. distance from the vaginal opening. Prostate discoid with about 25–30 alveolar glands. The Bathymetry. Intertidal to at least 30 m depth. vas deferens is moderate in length, penis long and twisted. Body length up to 19 mm. Remarks. Dendronotus albus has for almost 50 years persistently been confused with D. Distribution. Arctic – Kara Sea to Laptev Sea, robilliardi. Despite this, D. albus is readily Russia. distinguished from D. robilliardi by the number of dorsolateral appendages (four to Bathymetry. 10–20 m depth, occasionally º»ve in D. albus vs. º»ve to eight in D. robilliardi) deeper. and the number of dorsolateral appendages into which the digestive gland branches Remarks. This recently described species is penetrate (only in two anterior pairs in D. easily distinguished from its congenerics albus vs. three to six pairs including posterior (Korshunova et al., 2016a). This is a true Arctic ones in D. robilliardi). See Korshunova et al. species that so far has been recorded only (2016a) for discussion. from the Laptev Sea and Kara Sea.

Dendronotus arcticus Korshunova, Dendronotus bathyvela Martynov, Fujiwara, Sanamyan, Zimina, Fletcher & Martynov, Tsuchida, R. Nakano, N. Sanamyan, K. 2016 Sanamyan, Fletcher & Korshunova, 2020 Fig. 7 Fig. 7 Dendronotus arcticus Korshunova et al., Dendronotus bathyvela Martynov et al. 2020a: 2016a: 26–28, Figs 1, 3A. 507–510, Figs 4, 5.

Extended diagnosis. Body relatively narrow. Extended diagnosis. Body wide. Six to seven Five to six pairs of dorsolateral appendages. pairs of branched dorsolateral appendages. 10 Six to eight appendages of oral veil. Five to to 15 appendages of oral veil. Five appendages six appendages (apparently equal in size) of rhinophoral stalks. Lateral papilla of of rhinophoral stalks. Lateral papilla of rhinophoral sheaths absent. Rhinophores rhinophoral sheaths present. Rhinophores with 15–25 lamellae. Lip papillae 50–70. Basal with 15–18 lamellae. Lip papillae 15–25. Basal colour dull reddish-brownish with numerous colour brownish with scattered distinct distinct opaque white dots on notum, tips opaque white dots. Dorsal processes of jaws of lateral appendages, oral appendages. inclined posteriorly at approximately 55° Dorsal processes of jaws inclined posteriorly to the longitudinal axis of the jaw body and at approximately 52° to the longitudinal 0.47 of its length. Masticatory processes axis of the jaw body and 0.56 of its length. apparently bear indistinct denticles (may Masticatory processes apparently bear possess ridge-like structures). Radula with up indistinct denticles (may possess ridge-like to 39 rows of teeth. Central tooth with up to structures). Radula with up to 41 rows of teeth. 14 small denticles and reduced furrows. Up to Central tooth with high cusp and up to 35 nine lateral teeth with up to nine denticles. small distinct denticles without furrows. Up Ampulla folded twice. Bursa copulatrix is to 14 lateral teeth without or with few weak

 $        # $# !"%   %   ./-  |   . lateral denticles. Ampulla voluminous, with jaws with irregular denticles and possibly with approximately three loops. Bursa copulatrix ridge-like structures. Radula with up to 32 large, rounded to oval. Seminal receptaculum rows of teeth. Central tooth with deep furrows small placed distally at a short distance from and with up to 13 distinct denticles. Up to eight the vaginal opening. Prostate discoid with lateral teeth with up to ten denticles. Ampulla about 21–25 alveolar glands. The vas deferens with three loops. Prostate compact with about is long, penis elongate, narrow, curved. Body nine alveolar glands. Bursa copulatrix large, length up to 45 mm. rounded. Seminal receptaculum large placed distally at considerable distance from the Distribution. Paciº»c coast of Northern vaginal opening. The vas deferens is moderate Honshu, Japan, possibly may occur further in in length, penis relatively long, narrow. Body NW Paciº»c. length up to 18 mm.

Bathymetry. A shelf to deep-sea (bathyal) Distribution. Alarcón Rise, Gulf of California, species, so far recorded at depths of 249–510 Mexico, NE Paciº»c . m. Bathymetry. 2369 m depth. Remarks. This species has recently been described by Martynov et al. (2020a). Until Remarks. For this recently described deep-sea recently, only the single species D. robustus species there is no additional morphological was recognized among the wide-bodied, information and molecular data is very shallow-water dendronotids of the North limited. In particular, the dorsolateral Atlantic and the North Paciº»c shelf (Ekimova appendages need a further detailed study. et al., 2015). However, most recent data showed that “D. robustus” is a species complex, which Dendronotus comteti Valdés & Bouchet, 1998 was already separated into D. robustus s.str., Dendronotus comteti Valdés and Bouchet, D. velifer and D. bathyvela (Lundin et al., 2017; 1998: 320–323, Figs 1–3. Martynov et al., 2020a). Diagnosis (original description). Body narrow. Dendronotus claguei Valdés, Lundsten & Two or four pairs of partially branched, Wilson, 2018 partially unbranhed dorsolateral appendages. Dendronotus claguei Valdés, Lundsten, Wilson, Four to six appendages of oral veil. Four 2018: 412–415, Figs 3B, 9. appendages (similar in size) of rhinophoral stalks. Rhinophores with seven to eight Diagnosis (original description). Body very lamellae. Lateral rhinophoral papilla possibly narrow. Four to six pairs of dorsolateral absent. Lip papillae probably absent. Basal appendages (reported as unbranched). Four colour uniformly pale cream (preserved). appendages of oral veil. Four appendages of Jaws reported as devoid denticles. Radula rhinophoral stalks (innermost appendages with up to 28 rows of teeth. Central tooth from each side longest). Lateral papilla of with deep furrows and with up to ten distinct rhinophoral sheaths absent. Rhinophores denticles. Up to six lateral teeth with up to with 11 lamellae. Lip papillae not reported in º»ve denticles. The reproductive system was original description. Basal colour uniformly not studied in the original description. Body translucent white. Masticatory processes of length up to 5.5 mm (preserved).

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Distribution. Middle Atlantic, Lucky Strike denticles (may possess ridge-like structures). area in the Mid-Atlantic Ridge. Radula with up to 51 rows of teeth. Central tooth completely smooth in adults. Up to 16 Bathymetry. 1685 m depth. lateral teeth with up to 19 denticles. Ampulla relatively narrow with several (at least three) Remarks. This is the only known species loops. Bursa copulatrix very large, oval. that originated from deep-sea hydrothermal Seminal receptaculum small placed distally at habitats, and which was considered as a moderately short distance from the vaginal belonging to the genus Dendronotus according opening. Prostate discoid with up to 50 (and to its external and internal characters. more) alveolar glands. The vas deferens is Molecular data is necessary to determine its moderate in length, penis large, wide, with a phylogenetic position. blunt tip. Body length up to 14 cm.

Dendronotus dalli Bergh, 1879 Distribution. North Paciº»c (The Sea of Japan Fig. 7 in the west and at least Washington State, USA Dendronotus dalli Bergh, 1879: 94, Pl. 1, Fig. in the east to at least Chuckchii Sea; Russia, in 21; Pl. 2, Figs 9–12; Pl. 3, Figs 2–6; Robilliard, the Arctic (may possibly penetrate further to 1970: 452–455, Figs 13–15; McDonald, 2009: the west in the Arctic). 463 (bibliography); Martynov & Korshunova, 2011: 155–157 (veriº»ed photographs of living Bathymetry. Veriº»ed records from 5 to 50 m specimens, description and biology in NW depth. Paciº»c); Ekimova et al., 2015: 860–863, Figs 6C, 8G, 11, 12, 13A. Remarks. Until recently, the North Atlantic D. Dendronotus frondosus var. dalli Bergh, elegans (see below) has been confused with 1879 – Odhner, 1907: 19. the North Paciº»c D. dalli. The best external Not D. dalli sensu Bergh, 1886, Knipowitsch, feature to distinguish these two closely 1902, Roginskaya, 1987 (= D. elegans Verril, related species is the presence of an external 1880). opaque white pigment on the apical parts of the dorsolateral processes in D. dalli, whereas Extended diagnosis. Body relatively narrow. in D. elegans such a pigment, if present, is Four to eight pairs of branched dorsolateral internal. These species may co-occur in some appendages. Four to º»ve appendages of oral Arctic localities, but this needs to be further veil. Four to 12 appendages (equal in size or investigated. For D. dalli an absence of tertiary few posterior ones longer) of rhinophoral branches of the dorsolateral appendages stalks. Lateral papilla of rhinophoral sheaths was incorrectly mentioned by Ekimova et al. present. Rhinophores with 16–33 lamellae. (2015). There is a possibility that light-coloured Lip papillae 15–40 (and more). Basal colour varieties of D. kamchaticus were depicted uniform translucent gray, light yellow to under D. dalli, judging from photographs in for orange, brownish or pinkish, with external example Behrens (1980) (see also Korshunova opaque white pigment on tips of dorsolateral et al., 2016a). In some studies, Dendronotus appendages. Dorsal processes of jaws inclined dalli was incorrectly considered a synonym of posteriorly at approximately 60° to the D. frondosus (Odhner, 1936; MacFarland, 1966) longitudinal axis of the jaw body and 0.45 of its due to an underestimation of the importance length. Masticatory processes apparently bear of ontogenetic radula data (central teeth are

 $        # $# !"%   %   ./-  |   . denticulated in juveniles and usually smooth the jaw body and 0.43 of its length. Masticatory in adults). processes apparently bear denticles (which We have investigated a considerable num- may possess ridge-like structures). Radula ber of specimens and sequences of D. elegans with up to 42 rows of teeth. Central tooth from the North Atlantic and neighbouring completely smooth in adults. Up to at least subarctic regions. There are no conº»rmed 13 lateral teeth with up to eight denticles records of true D. dalli in the North Atlantic (can be more). Bursa copulatrix large, oval. and neighbouring regions. The indication of Seminal receptaculum small placed distally at D. dalli from the North Atlantic (Genbank a moderately short distance from the vaginal Û AF249800) in a table of the paper of opening. Prostate discoid with up to 60 and Wollscheid-Lengeling et al. (2001) is due to a more alveolar glands (their number was mix-up of localities, in this case with the North considerably underestimated in the original Paciº»c. The same can be said about Cadlina description of its new synonym D. niveus (see luteomarginata from the North Atlantic in remarks below). The vas deferens is moderate the same paper (Wollscheid-Lengeling et al., in length, penis strong, conical. Body length 2001) for which Johnson (2010: 142) explained up to 70 mm. that they originated from the North Paciº»c. Distribution. North Atlantic (known southern Dendronotus elegans Verrill, 1880 limit – Cape Cod, USA) to the Arctic Figs 7, 8 (including Barents Sea and White Sea, known Dendronotus elegans Verrill, 1880: 385–386. northeastern limit is the Laptev Sea, Russia). Dendronotus “elongatus”: Roginskaya, 1987: 175 (lapsus calami for D. elegans Verrill, 1880). Bathymetry. 10–25 m (White Sea) to 258 m Dendronotus niveus Ekimova, Korshunova, depth (western North Atlantic). Shepetov, Neretina, Sanamyan & Martynov, 2015: 864–869, Figs 6D, 8F, 13B, 14, 15 syn. nov.; Remarks. Ekimova et al. (2015) overlooked Valdés et al., 2017: 6–7, 4B, 5B. important details in the original description Dendronotus dalli sensu Bergh, 1886 (non of D. elegans Verrill, 1880, for which a big Bergh, 1879); Robilliard, 1970: 452; Bleakney, with uniform salmon-coloured body 1996: 109 (D. elegans incorrectly mentioned as was used, and a special mentioning was synonym of D. dalli). made of the completely smooth central teeth (Verrill, 1880: 385–386). These characters are Extended diagnosis. Body relatively narrow. identical to those of the recently described Five to ten pairs of branched dorsolateral D. niveus Ekimova et al., 2015. The available appendages. Six to nine appendages of oral molecular data (Valdés et al., 2017; present veil. Four to º»ve appendages (equal in size study, º»gs 1–3) also show that specimens or few posterior ones longer) of rhinophoral identiº»ed as D. niveus from o¼f the east coast stalks. Lateral papilla of rhinophoral sheaths of (relatively close to the present. Rhinophores with 8–23 lamellae. Lip type locality of D. elegans) are identical to papillae 8–50. Basal colour uniform, pinkish specimens from the type locality of D. niveus to light orange, with internal opaque white in the White Sea. The holotype of D. elegans pigment of tips of dorsolateral appendages. in the Smithsonian Institution (NMNH, 2020) Dorsal processes of jaws inclined posteriorly at is not suitable for molecular study. However, approximately 40° to the longitudinal axis of a microslide with radula of the holotype is

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

ë À 8 Dendronotus elegans. External and internal morphology. a: living specimen ZMMU Op-713 from Greenland, near East Atlantic North American coast; b: jaw, same specimen; c; posterior part of radula, general overview, same specimen; d: close up of smooth central teeth, same; e: close up of lateral teeth, same; f: living specimen ZMMU Op-714 (as “D. niveus”) from the White Sea (subarctic); g: jaw, same specimen; h: masticatory edge of the jaws, same; i: posterior part of radula, general overview, same specimen; j: close up of smooth central teeth, same; k: living specimen ZMMU Op-715 from the Laptev Sea (Arctic); l: jaw, same specimen; m: posterior part of radula, close up of smooth central teeth, same; n: lateral teeth, same. o: discoid prostate with numerous alveols (which considerably exceed the considerably underestimated number presented in the º»rst description of D. niveus), same specimen; Scale bars: b, l, 1mm; c: 300 µm; d, h–j, m: 100 µm; g: 1000 µm; e, n: 30 µm. ÚÀÚ Á . × Ø  (, )  .  (ë). Ø Ø ÛÀÚ Á . ØÁ

 $        # $# !"%   %   ./-  |   . kept separately in the Yale Peabody Museum generally be described as highly branched, of Natural History (YPM IZ 010761.GP) and giving this species a ÐÑu¼fy appearance. Eight to available as online images (Yale Peabody 15 appendages of oral veil. Five to six appendages Museum, 2020), which conº»rm the presence (posterior one very long) of rhinophoral stalks. of massive smooth central teeth. The date Lateral papilla of rhinophoral sheaths present. (6 September 1879), the depth (26 fathoms = 48 Rhinophores with 11–17 lamellae. Lip papillae meters) and locations (Stellwagen Bank, o¼f 20–49. Basal colour reddish-brown with opaque Cape Cod) for collecting data of the holotype white stripes between dorsolateral processes to of D. elegans exactly match radula YPM IZ uniform light yellow with small brown spots. 010761.GP, the preserved body USNM 842116 Dorsal processes of jaws inclined posteriorly and the original description of D. elegans at approximately 60° to the longitudinal in Verrill (1880). Additional morphological axis of the jaw body and 0.43 of its length. information of the molecularly investigated Masticatory processes apparently with º»ne specimens of D. elegans of the present study denticles. Radula with up to 37 rows of teeth. is given in º»g. 8. It is evident that specimens Central tooth with up to 20 small denticles from the Atlantic North American coast without furrows, rarely completely smooth. and from the Arctic Eurasian regions are Up to 11 (usually 9–10) lateral teeth with up morphologically similar according to their to seven denticles. Ampulla with at least two external and internal features. We therefore loops. Bursa copulatrix large, rounded. Seminal synonymize Dendronotus niveus Ekimova et receptaculum small placed distally at a short al., 2015 with D. elegans Verrill, 1880. Another distance from the vaginal opening. Prostate species with smooth central teeth at the discoid with a range of 35–40 alveolar glands. adult stage is D. dalli Bergh, 1879 (see above, The vas deferens is very long, penis relatively º»gs 1, 2), but this one is found predominantly narrow, bent. Body length up to 100 mm. in the North Paciº»c and di¼fers from D. elegans in habitus morphology (apical parts Distribution. NE Atlantic, so far known from of dorsolateral appendages with external the Netherlands, Norway, Sweden and the UK. opaque white pigment vs. with internal or absent white pigment in D. elegans) and Bathymetry. 15–150 m depth. molecular data. Remarks. This species has been separated Dendronotus europaeus Korshunova, from from D. frondosus and D. lacteus based on Martynov, Bakken & Picton, 2017 both morphological and molecular evidence Fig. 7 (Korshunova et al., 2017b). Dendronotus europaeus Korshunova, Martynov, Bakken, Picton, 2017: 1–8, Figs. 1–2; Dendronotus frondosus (Ascanius, 1774) Supplementary information, text and º»gure. Fig. 7 Dendronotus frondosus sensu Thompson Amphitrite frondosa Ascanius, 1774: 155–158, and Brown, 1984 (partim.); not Ascanius, 1774. Pl. 5, Fig. 2 (neotype selected in Ekimova et al. 2015). Extended diagnosis. Body relatively narrow. Dendronotus frondosus (Ascanius, 1774) – Six to nine pairs of branched dorsolateral Odhner, 1936: 1105–1109, Fig. 39 (mixture of appendages, including two to three posterior several species); Robilliard, 1970: 441–446, ones. The dorsolateral appendages can Pl. 63, Fig. 29, text and Figs 4, 7, 8, 9 (partim,

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   mostly referred to D. venustus); Ekimova et al. eight) lateral teeth with up to seven denticles. 2015: 848–857, Figs 1–5, 6A, 7A, 8A. Ampulla voluminous, folded. Bursa copulatrix arborescens Müller, 1776: 229 (mixture large, oval to rounded. Seminal receptaculum of several species). small placed distally at a moderately short Doris cervina Gmelin, 1791: 3105, no. 12. distance from the vaginal opening. Prostate Tritonia reynoldsii Couthouy, 1838: 74–80, discoid with range 16–30 alveolar glands. Pl. 2, Figs 1, 2, 3, 4. The vas deferens is moderate in length, penis Tritonia felina Alder & Hancock, 1842: 33. relatively long, curved. Common body length Tritonia pulchella Alder & Hancock, 1842: no more than 50 mm. 33–34. Tritonia ascanii Møller, 1842: 5. Distribution. North Atlantic, both east and Amphitridea fabricii Kröyer, 1847: 114. west parts, not distributed to the Arctic Campaspe pusilla Bergh, 1863: 471–478, Pl. further than the easternmost border of the 12, Figs 28–35. Barents Sea. Campaspe major Bergh, 1886: 21–24; Pl. 1, Figs. 23–26; Pl. 2, Figs 1–11. Bathymetry. Intertidal to 20–30 m depth. Dendronotus luteolus Lafont, 1871: 267. Dendronotus junior Mörch, 1875: 125. Remarks. Three common North Atlantic Dendronotus arborescens var. aurantiaca shallow water species, namely D. europaeus Friele, 1879: 284. Korshunova et al., 2017b, D. frondosus : Mixture of several species under the (Ascanius, 1774), and D. lacteus, show an name Dendronotus frondosus in the literature extremely similar range of habitus variation before 2015. (see Korshunova et al., 2017b) and have been confused in older literature. Therefore, the Extended diagnosis. Body narrow. Five to six veriº»ed distribution of D. frondosus can only pairs of branched dorsolateral appendages. be assessed using recent data (Korshunova Four to seven appendages of oral veil. Four to et al., 2017b) that covers at least the North East º»ve appendages (middle and posterior ones Atlantic from Norway and also the Barents can be longer) of rhinophoral stalks. Lateral Sea and the White Sea to possibly France/ papilla of rhinophoral sheaths present. northern Spain. The range of D. frondosus Rhinophores with six to 12 lamellae. Lip does not extend further than the the true papillae four to 12. Basal colour brownish to Arctic. Also, several deep-water records of reddish-brown, often with small white and D. frondosus (e.g., Odhner, 1939; Swennen, yellow specks, but usually without opaque 1961; Thompson & Brown, 1984) most likely white stripes between dorsolateral processes refer to D. lacteus, since according to our data to completely white translucent specimens. D. frondosus was usually found shallower Dorsal processes of jaws inclined posteriorly than 20–30 m depth. We have examined an at approximately 60° to the longitudinal extensive collection of various Dendronotus axis of the jaw body and 0.4 of its length. from the NW Paciº»c and NE Paciº»c and Masticatory processes apparently bear ridge- have so far not found any specimen from the like structures and denticles. Radula with up North Paciº»c for which the identity could to 42 rows of teeth. Central tooth with deep be conº»rmed as D. frondosus despite the furrows and with up to 14 (common range presence of a few published records (e.g., 8–12) distinct denticles. Up to 10 (usually up to Ekimova et al., 2016). In this respect, the

 $        # $# !"%   %   ./-  |   . common occurrence of true D. frondosus in 1–3) with stable distinct di¼ferences in exter- the North Atlantic and the extreme rarity nal and internal morphology (º»g. 4) and of D. frondosus in the North Paciº»c, with without any gene ÐÑow between each other. the simultaneous common presence in Dendronotus kalikal thus does not occur the North Paciº»c of other species formerly naturally in the North Atlantic. A potential considered as “D. frondosus”, (e.g., D. kalikal, future introduction of some North Paciº»c D. kamchaticus, D. primorjensis, D. venustus), species from the “D. frondosus megacom- clearly suggests that true D. frondosus does plex” may take place, but so far there are not occur naturally in the North Paciº»c. no reliable records of any North Paciº»c spe- Here, it is largely substituted by other species cies in the North Atlantic. At least recently, from the D. frondosus species complex. In a case of evident anthropogenic transpor- particularly, D. primorjensis endemic to the tation of the dendronotid Pseudobornella Sea of Japan, is morphologically similar to orientalis was reported from Japan or China D. frondosus and sister to it according to our to NE Paciº»c (Agarwal et al., 2017) (for tax- molecular phylogenetic analysis (º»g. 1). D. onomy of Pseudobornella see below). Thus, primorjensis has several traceable diagnostic most previous records of this species from features in radula and reproductive system the Northwest Paciº»c refer to one of three (see Korshunova 2016a and Synopsis below). recently described species, D. kamchaticus By such features and molecular data D. Ekimova et al., 2015, D. kalikal Ekimova et primorjensis could be excellent candidate al., 2015, or D. primorjensis Martynov et al., for an example of a relict species, with 2015, and in the Northeast Paciº»c, previous close relatives with ranges extending to the records refer to D. venustus Stout et al., 2010 Arctic, which afterwards formed a separate or D. kamchaticus Ekimova et al., 2015 (see species, D. frondosus, in the North Atlantic. also Korshunova et al., 2016a, b). As a result of such an evolutionary history, D. primorjensis and D. frondosus are evidently Dendronotus gracilis Baba, 1949 naturally separated from from closely related Dendronotus gracilis Baba, 1949: 167, Pl. 35, North Paciº»c and North Atlantic species, Fig. 127, text-º»g. 109; Robilliard, 1970: 461–462; because true D. frondosus does not occur in Nakano, 2018: 385. the Arctic. On the contrary, the potentially disrupted populations of D. frondosus in Diagnosis (original description). Body the North Atlantic and the North Paciº»c do narrow. Four pairs of branched dorsolateral not reveal morphological and molecular appendages. Four veil appendages. Five to di¼ferences. Most reliable explanations for six processeses of the rhinophoral stalks. such an occurrence of few of so far known Rhinophores with 18 lamellae. Basal colour specimens of D. frondosus in the North bluish-white, with numerous scattered Paciº»c can be an anthropogenic introduction yellow spots, various processes opaque white. or a mistake in the sorting or processing of Masticatory processes of jaws with denticles. collection material. Dendronotus frondosus Radula with up to 41 rows of teeth. Central does not occur naturally in the North Paciº»c. tooth with up to 20 small denticles without In this study we also evidently show that furrows, rarely completely smooth. Up to the North Paciº»c D. kalikal and the North eight lateral teeth with up to nine denticles. Atlantic and Arctic D. yrjargul sp. nov. are Reproductive system unknown. Body length invariably placed in two distinct clades (º»gs 25 mm (living, original description).

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Distribution. Japan, Paciº»c side of Honshu, transportation, which is hard to explain in this Japan. case and needs veriº»cation.

Bathymetry. 160 m depth (original description). Cooper, 1863 Fig. 7 Remarks. This species is insu¼º»ciently known. Dendronotus iris Cooper, 1863: 59; Information on the internal morphology of MacFarland, 1966: 257–265, Pl. 47, Figs 12– specimens from the type locality is restricted 18; Plate 48, Figs 1–6; Pl. 49, Fig. 4; Pl. 50, Fig. to the original description (Baba, 1949). 1; Pl. 51, Figs 1–5; Robilliard, 1970: 446–450, Additional data in Pola and Stout (2008) Figs 4–6, 10–12, Pl. 63, Fig. 30; Behrens, 1980: referring to D. gracilis specimens are from 74; Behrens & Hermosillo, 2005: 94. considerably remote locations o¼f the tropical Dendronotus giganteus O’Donoghue, 1921: island of Okinawa, whereas the type locality 187–190; Pl. 4, Fig. 47; Pl. 5, Figs 57–59. of D. gracilis is in the temperate Sagami Bay. Their study lacks molecular data and may Extended diagnosis. Body relatively represent a separate species. The Okinawan narrow. Four to eight pairs of branched specimen indicated in Pola and Stout (2008) dorsolateral appendages. Two to four oral has eight pairs of dorsolateral processes, six veil appendages. Two to five appendages processess of the oral veil, and 12 rhinophoral (longest internal) of rhinophoral stalks. lamellae, whereas according to the original Lateral papilla of rhinophoral sheaths description in Baba (1949) there are four pairs present. Rhinophores with 15–31 lamellae. of dorsolateral processes, four processesses 20–40 lip papillae. Basal colour grayish to of the oral veil, and 18 rhinophoral lamellae, reddish-orange, with white line bordered respectively. Taking into consideration that foot edge. Dorsal processes of jaws inclined details of colour patterms between the original posteriorly at approximately 40° to the description of D. gracilis and Okinawan longitudinal axis of the jaw body and 0.35 specimens are also di¼ferent as indicated by of its length. Masticatory processes bear Pola and Stout (2008: 65), we cannot conº»rm denticles (possibly including ridge-like the conspeciº»city of the Okinawan specimens structures and denticles). Radula with up with the original D. gracilis, even if we consider to 61 rows of teeth. Central tooth quite the potentially cooler waters in Okinawa at 69 narrow, with deep furrows and with up to m depth and that only a few specimens were 18 distinct denticles. Up to 21 lateral teeth involved in the comparison. Furthermore, Pola commonly smooth, occasionally with up to and Stout (2008) considered specimens from eight denticles. Ampulla relatively narrow New Zealand as belonging to D. gracilis as well, with numerous loops. Bursa copulatrix which also show di¼ferences in colouration. moderate in size, oval, bent. Seminal However, in the light of a modern emerging receptaculum small, placed distally at a paradigm of multilevel organismal diversity moderately short distance from the vaginal that challenges the previously concept of opening. Prostate very long, non-discoid, widely claimed polytypic species with broad linear, with numerous alveolar glands. The ranges in favour of considerably smaller, often vas deferens is short, penis moderately geographically restricted species, a potential short, straight, with blunt tip. Body length presence of true D. gracilis in New Zealand of live specimens may reach more than can only be the result of an anthropogenic 20 cm.

 $        # $# !"%   %   ./-  |   .

Distribution. NE Paciº»c, from southern at approximately 56° to the longitudinal Alaska to California (but records from Baja axis of the jaw body and 0.47 of its length. California need to be conº»rmed). As an adult Masticatory processess with ridge-like it feeds on cerianthid anthozoans (Wobber, structures. Radula with up to 38 rows of teeth. 1970; Shaw, 1991), quite di¼ferent to other Central tooth with up to 17 small, distinct Dendronotus species. denticles with furrows. Up to eight lateral teeth with up to nine denticles. Ampulla Bathymetry. 10–30 m depth. moderately narrow, with several loops. Bursa copulatrix large, oval. Seminal receptaculum Remarks. This species is clearly distinguished small, placed distally at a moderately short by a combination of external characters distance from the vaginal opening. Prostate (uniform grayish to reddish-orange colour, discoid with 50–60 alveolar glands. The vas massive dorsal appendages) and internal deferens is moderate in length, penis conical, characters (linear instead of discoid relatively thin. Body length up to 21.5 mm. prostate – unique for the genus), and also by adult behavioral patterns (special “rearing Distribution. Paciº»c coast of Northern Honshu, attack” feeding on species of Ceriantharia, Japan. e.g., Shaw, 1991). Dendronotus nanus Marcus et Marcus, 1967 was considered a synonym Bathymetry. A deep-sea (bathyal) species, of D. iris (Stout et al., 2010), but in the so far recorded at 670–761 m depth (for absence of molecular data and presence of discussion of the di¼ferent usage of the term morphological di¼ferences (see below) we “bathyal”, see Rybakova et al., 2020). prefer to retain D. nanus as a separate species (see details below). Dendronotus iris possibly Remarks. This species is recently described in is the largest species of its genus. detail (Martynov et al., 2020a).

Dendronotus jamsteci Martynov, Fujiwara, Dendronotus kalikal Ekimova, Korshunova, Tsuchida, R. Nakano, N. Sanamyan, Shepetov, Neretina, Sanamyan & Martynov, K. Sanamyan, Fletcher & Korshunova, 2020 2015 Fig. 7 Fig. 7 Dendronotus jamsteci Martynov, Fujiwara, Dendronotus kalikal Ekimova, Korshunova, Tsuchida, Nakano, Sanamyan, Sanamyan, Shepetov, Neretina, Sanamyan & Martynov, Fletcher, Korshunova, 2020a: 500–505, Figs 2, 2015: 872–874, 6F, 8E, 16C, 18B, 19; Korshunova 5A. et al., 2019b: 8–9, Figs 1, 3M–Q.

Extended diagnosis. Body narrow. Six to seven Extended diagnosis. Body narrow. Four to º»ve pairs of branched dorsolateral appendages. pairs of branched dorsolateral appendages Six to nine appendages of oral veil. Four to (plus 2–3 posterior smaller processes). Four º»ve appendages (posterior ones longest) to º»ve oral veil appendages. Five appendages of rhinophoral stalks. Lateral papilla of (apparently similar in size) of rhinophoral rhinophoral sheaths present. Rhinophores stalks. Lateral papilla of rhinophoral sheaths with 12–15 lamellae, 20–25 lip papillae. Basal present. Rhinophores with 8–14 lamellae. Four colour whitish with brownish spots and streak. to six lip papillae. Basal colour brownish, with Dorsal processes of jaws inclined posteriorly creamy to whitish areas, dorsum, dorsolateral

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   appendages, and upper sides of foot bear & Martynov, 2015: 869–872, Figs 6E, 8D, 16A, brownish orange lines and dots, partly arranged B, 17, 18A; Korshunova et al., 2016a: 32–36, Figs in blurred subparallel pattern. Dorsal processes 4C, D, 5. of jaws inclined posteriorly at approximately 35–40° to the longitudinal axis of the jaw body Extended diagnosis. Body narrow. Five to six and 0.45 of its length. Masticatory processes pairs of branched dorsolateral appendages. bear denticles (possibly including ridge-like Four to six oral veil appendages. Five to structures). Radula with up to 31 rows of teeth. six appendages (apparently similar in size Central tooth with deep furrows and with up or posterior longer) of rhinophoral stalks. to 18 distinct denticles. Up to eight lateral teeth Lateral papilla of rhinophoral sheaths with up to eight denticles. Ampulla moderately present. Rhinophores with 5–11 lamellae. Lip voluminous, folded. Bursa copulatrix large, papillae 4–15. Basal colour semitransparent considerably elongate, narrowly-oval. Seminal to brownish with irregularly scattered small receptaculum small placed distally at a short brownish dots. Dorsal processes of jaws distance from the vaginal opening. Prostate inclined posteriorly at approximately 70° to discoid with circa 20 alveolar glands. The vas the longitudinal axis of the jaw body and 0.37 deferens is moderate in length, penis relatively of its length. Masticatory processes possibly long, slightly curved. Body length up to 15 mm bear denticles and ridge-like structures. Radula (preserved). with up to 44 rows of teeth. Central tooth with almost completely reduced denticles and Distribution. NW Paciº»c from Kamchatka furrows at adult stages, subadults may bear to northern Kurile Island and Bering Strait, small denticles. Up to eleven lateral teeth with Russia. up to nine denticles. Ampulla moderately voluminous with several (at least two or Bathymetry. 14–60 m depth. three) folded loops. Bursa copulatrix large, rounded. Seminal receptaculum small, placed Remarks. Although this species was previously distally at a moderately short distance from confused with D. frondosus, D. kalikal is very the vaginal opening. Prostate discoid with distinct from it by morphological features circa 7–25 alveolar glands. The vas deferens (mottled body, brownish colouration with is moderate in length, penis relatively long, indistinct but commonly present dorsal slightly curved. Body length up to 30 mm. blurred subparallel stripes), radular patterns and molecular phylogenetic data. In the Distribution. NW Paciº»c from Kamchatka to present study we provide evidence that in the Japan Sea, Russia, and NE Paciº»c in Salish Sea, North Atlantic to the Kara Sea there is a sister Canada and USA. species, D. yrjargul sp. nov., separate from D. kalikal, see above. Bathymetry. 7–17 m depth.

Dendronotus kamchaticus Ekimova, Remarks. Though this species previously was Korshunova, Shepetov, Neretina, Sanamyan confused with D. frondosus, D. albus and D. & Martynov, 2015 venustus, D. kamchaticus is well delineated Fig. 7 from those species according to the Dendronotus kamchaticus Ekimova, morphological features (semitransparent Korshunova, Shepetov, Neretina, Sanamyan grayish with white-tipped dorsolateral

 $        # $# !"%   %   ./-  |   . appendages and digestive gland often Dendronotus lacteus (Thompson, 1840) penetrating all branches of all dorsolateral Fig. 7 appendages to body brownish colouration Tritonia lactea Thompson, 1840: 88–89, Pl. without dorsal subparallel stripes), radular 2, Fig. 3. patterns (usually almost smooth central teeth Dendronotus lacteus Eliot, 1910: 161; in adults) and molecular phylogenetic data Nordsieck, 1972: 68; Pruvot-Fol, 1954: 356; (º»gs 1, 2). In the original description branches of Thollesson, 1998: 191–193, Figs 1–3; Evertsen the dorsolateral appendages were incorrectly and Bakken, 2005: 18; Korshunova et al., 2017: described as bulbous and short (see Ekimova 2–7, Figs 1, 2. et al., 2015: Fig. 8D), despite the presence of not Becher, 1886: 14. this species in the same paper (Ekimova et al., not Eliot, 1910:112 (= D. dalli). 2015: Fig. 16A, B) represented by photos of live = D. frondosus auct. non Ascanius, 1774. paratypes with more elongated, not exactly bulbous, though still shortened branches. Extended diagnosis. Body relatively narrow. Further, in a specimen of D. kamchaticus from Four to eight pairs of branched dorsolateral the Sea of Japan the branches were described appendages which in adults commonly show as elongated and also very di¼ferent from the speciº»c “curly” pattern though other specimens original description and Fig. 8D in Ekimova with more “typical” for genus Dendronotus (2015). This signiº»cant discrepancy compared elongated appendages may occur. Four to ten with the original description was explained: oral veil appendages. Four to six appendages “because di¼ferent ontogenetic stages were (inetrnal middle and posterior ones longest) used in the description and this study, as of rhinophoral stalks. Lateral papilla of well as cerata morphology may be altered in rhinophoral sheaths present. Rhinophores preserved specimens” (Ekimova et al., 2016: with 10–14 lamellae. At least more than 15 lip 38). However, this discrepancy was not due papillae. Basal colour from uniformly dark red to to the di¼ferent ontogenetic stages. In the uniform white, often with small brownish dots. original description of D. kamchaticus, the Redddish and brownish specimens may have type specimens were represented by an almost white stripes between cerata, white specimens adult holotype (12 mm preserved length) milky, non-transparent in appearance. Dorsal with a mature reproductive system whereas processes of jaws inclined posteriorly at seven paratypes were subadults (2–10 mm approximately 55–60° to the longitudinal of preserved length). Most importantly, the axis of the jaw body and 0.52 of its length. drawing in Ekimova (2015: Fig. 8D) depicted the Masticatory processes possibly bear denticles largest holotype, with an adult reproductive and ridge-like structures. Radula with up to 43 system and the artiº»cially bulbous branches rows of teeth. Central tooth with up to 40–50 were only seen in the preserved specimen, small denticles (commonly smaller), often without any considerable ontogenetic with peculiar ribs, never completely smooth. modiº»cations. The largest specimen so far Up to 16 lateral teeth with up to º»ve denticles. known of D. kamchatcius (adult, 30 mm) Ampulla moderately narrow with several from the NE Paciº»c, also has considerably windings or folded loops. Bursa copulatrix elongated branches (Korshunova et al., large, rounded to oval. Seminal receptaculum 2016a). This fully conº»rms that the branches small placed distally at considerable distance were described and º»gured incorrectly in the from the vaginal opening. Prostate discoid with original description. a range from 12 to more than 30 alveolar glands.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

The vas deferens is relatively long in length, Ampulla was not described in the original penis relatively long, bent. Body length up to description. Bursa copulatrix large, rounded 150 mm. to oval. Seminal receptaculum small placed distally at considerable distance from the Distribution. North Atlantic and Arctic vaginal opening. Glandular part of prostate (eastern limit at least Laptev Sea, Russia). was reported in the original description as in early development of differentiation. Bathymetry. From circa 5 to 427 m depth; The vas deferens is extremely long, with common at shallow, subtidal depths at around numerous loops, penis moderately short, 10–20 m. straight, with blunt tip. Body length up to 13 mm. Remarks. Dendronotus lacteus together with D. frondosus and D. europaeus form a species Distribution. NE Paciº»c, Baja California, complex that is di¼º»cult to distinguish Puerto Peñasco, Sonora, Mexico. based on external characteristics, but in which it is still possible to º»nd º»ne-scale Bathymetry. Floating on seawater surface. morphological di¼ferences. In contrast, the radulae of these three species demonstrate Remarks. Dendronotus nanus was considered distinctly di¼ferent patterns (Korshunova et a synonym of D. iris by Stout et al. (2010), but al., 2017b). in the absence of molecular data we prefer to keep them separate. Furthermore, despite the Dendronotus nanus Marcus & Marcus, 1967 fact that Stout et al. (2010) mentioned only Dendronotus nanus Marcus & Marcus, minor morphological di¼ferences between 1967: 210–214, Figs 64–65; Farmer, 1980: 72; D. iris and D. nanus, Marcus and Marcus Robilliard, 1972: 428–430. (1967) in their original description specially highlighted that D. nanus di¼fers from D. iris Diagnosis (original description). Body by a considerably longer vas deferens in narrow. Six pairs of branched dorsolateral of only 13 mm length (an considered appendages. Circa six oral veil appendages. to be not fully mature), whereas D. iris that Five appendages (posterior ones the longest) was studied anatomically by Robilliard of rhinophoral stalks. Lateral papilla of (1970) had a considerably larger average body rhinophoral sheaths present. Rhinophores length 65–120 mm but a signiº»cantly shorter with about 15 lamellae. Circa 20 lip papillae. vas deferens. In the BOLD database there is a Basal colour semitransparent grayish with sequence of D. nanus (in GenBank identiº»ed a slight reddish hue and a little brown as D. iris), but from a very remote location in pigment, with white line bordered foot British Columbia whereas the type locality edge, white tipped dorsolateral appendages of D. nanus is Baja California, Sonora, thus with blackish brown subapical rings and an this available sequence does not represent orange base. Masticatory processeses with a true D. nanus. Even if taken into account crested pointed denticles. Radula with up to that the glandular part of vas deferens 26 rows of teeth. Central tooth quite narrow, likely is underdeveloped in D. nanus (see with deep furrows and with up to 10 distinct Marcus & Marcus, 1967), the non-glandular denticles. Up to 14 lateral teeth commonly muscular part of vas deference of D. nanus is smooth, occasionally with up to 12 denticles. considerably longer than in D. iris. Robilliard

 $        # $# !"%   %   ./-  |   .

(1972) compared D. iris and D. nanus in detail Dendronotus primorjensis Martynov, and mentioned similarity of both species Sanamyan & Korshunova, 2015 but did not synonymize them. All these Fig. 7 characters need to be re-analysed before Dendronotus primorjensis Martynov et al., any further attempts at synonymizing these 2015a: 60, Fig. 5A–G; Martynov et al., 2015c: species should be done. 75–76; Korshunova et al., 2016b: 15–28, Figs 1, 2; Korshunova et al., 2016a: 20–32; Korshunova Dendronotus patricki Stout, Wilson & et al., 2019b: 6–8, Fig. 3A–F. Valdés, 2011 Dendronotus dudkai Ekimova et al. (2016): Dendronotus patricki Stout et al., 2011: 63–67, 33–38, Figs 9–11 (synonym of D. primorjensis). Figs 2–4. Extended diagnosis. Body narrow. Five to nine Diagnosis (original description). Body pairs of branched dorsolateral appendages broad. Three to four pairs of branched (including 2–3 smaller pairs). Five to twelve dorsolateral appendages. Six appendages of oral veil appendages. Four to º»ve appendages oral veil. Four appendages (posterior ones (internal middle and posterior ones longest) of long) of rhinophoral stalks. Lateral papilla rhinophoral stalks. Lateral papilla of rhinophoral of rhinophoral sheaths absent. Number of sheaths present. Rhinophores with 8–12 rhinophoral lamellae and lip papillae not lamellae. Five to 12 lip papillae. Basal colour non- indicated in the original description. Colour uniformly reddish brown with few opaque white uniformly reddish brown (appears in original stripes between dorsolateral processes to more description as rather pinkish) with a semi- uniformly olive almost without white pigment, transparent body. Masticatory processeses or almost lacking general pigmentation. with ridge-like structures. Radula with up to Dorsal processes of jaws inclined posteriorly at 31 rows of teeth. Central tooth with up to 20 approximately 45–47° to the longitudinal axis of small denticles without furrows. Up to eight the jaw body and 0.48 of its length. Masticatory lateral teeth without denticles. Ampulla processes bear denticles and ridge-like moderately voluminous with two folded structures. Radula with up to 41 rows of teeth. loops. Bursa copulatrix very large, rounded Central tooth with deep furrows and with up to to oval. Seminal receptaculum moderate in 18 (common range 12–14) distinct denticles. Up size placed distally near the vaginal opening. to nine lateral teeth with up to eight denticles. Prostate discoid with range at least 10 alveolar Ampulla moderately narrow with several glands. The vas deferens is very long, penis windings or folded loops. Bursa copulatrix large, relatively long, straight, conical. Body length rounded. Seminal receptaculum small placed up to 25 mm. distally at a moderately short distance from the vaginal opening. Prostate discoid with range 12– Distribution. Whalefall ‘Patrick’ area (36º42’ 19 alveolar glands. The vas deferens is moderate 30.31’’ N, 122º 06’ 18.72’’W), Monterey Canyon, in length, penis relatvely long, curved. Reported USA. body length up to 35 mm.

Bathymetry. 1819–1822 m depth. Distribution. NW Paciº»c, common in the northern part of the Sea of Japan, at both the Remarks. So far only known from three Russian coast and at Hokkaido on the Japanese specimens collected at the type locality. coast (Korshunova et al., 2016b, 2019b).

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Bathymetry. 3–20 m depth. teeth usually have more distinct denticles. We therefore retain the North Paciº»c D. purpureus Remarks. Dendronotus dudkai Ekimova et al. as a separate species, potentially a sister species (2016) is a synonym of D. primorjensis according to D. lacteus, until further studies are done. to morphological, molecular and geographical data. For discussion see Korshunova et al. Dendronotus robilliardi Korshunova, (2016b, 2019b). Sanamyan, Zimina, Fletcher & Martynov, 2016 Fig. 7 Dendronotus purpureus Bergh, 1879 Dendronotus robilliardi Korshunova, Dendronotus purpureus Bergh, 1879: 89–93, Pl. Sanamyan, Zimina, Fletcher & Martynov, I, Figs 18–20; Pl. 3, Figs 7–12.; Bergh, 1904:15; 2016: 28–32, Figs 2, 3, 5. MacFarland, 1966: 254; Odhner, 1936:1108. Non Dendronotus purpureus var. auranti- Extended diagnosis. Body narrow. Five to eight aca Friele, 1879 (= D. frondosus or D. lacteus). pairs (commonly more than six) of branched dorsolateral appendages, digestive gland Diagnosis (original description). Exact penetrates at least three pairs of dorolateral number of pairs of branched dorsolateral appendages. Four to six appendages of oral appendages unknown (at least, more than veil. Four to six appendages of rhinophoral three). Living colour purple, preserved reddish stalks. Lateral papilla of rhinophoral sheaths brown. Circa seven oral veil appendages. present. Rhinophores with 11–15 lamellae. Five appendages (described as equal in size, Five to ten lip papillae. Basal colour white, likely due to preservation) of rhinophoral dorsolateral appendage variable opaque stalks. Lateral papilla of rhinophoral sheaths orange or opaque white, including tip. Dorsal present. Rhinophores with circa 20 lamellae. processes of jaws inclined posteriorly at Several lip papillae (no exact count present in approximately 50–60° to the longitudinal original description). Masticatory processes axis of the jaw body and 0.43–0.45 of its of jaws with small denticles. Radula with up length. Radula with up to 43 rows of teeth. to 44 rows of teeth. Central tooth with a very Masticatory processes with ridge-like º»ne denticulation. Up to 14 lateral teeth with structures and smaller denticles. Central up to six denticles. Prostate and vas deferens tooth with up to 17 small, distinct denticles unknown, penis long, straight, conical. Body without furrows. Up to nine lateral teeth with length up to 20 mm. up to eight denticles. Ampulla is hook-shaped to swollen kidney-shaped, with various Distribution. North Paciº»c, . Port degrees of folding. Bursa copulatrix large, Moller, Alaska, USA. rounded, with small seminal receptaculum placed distally at a short distance from the Bathymetry. 31 m depth. vaginal opening. Prostate discoid with about 19–30 alveolar glands. The vas deferens is Remarks. According to its original description short, penis relatively short, curved. Body (Bergh, 1879), D. purpureus has 14 lateral radula length up to 35 mm. teeth and is thus similar in this respect to the North Atlantic and Arctic species D. lacteus. Distribution. NW (from South Korea to However, D. lacteus of such a size has usually Commander Islands) and NE Paciº»c (from less than 14 lateral teeth, and also the central Alaska to at least Washington State, Canada

 $        # $# !"%   %   ./-  |   . and USA). Northern Japan (photographic range does not extend further eastward than record only, Korshunova et al., 2019b). the Barents Sea.

Depth. circa 5–30 m. Bathymetry. Intertidal to less than 20 m deep.

Remarks. See Remarks for D. albus. Remarks. T.A. Korshunova’s and A.V. Martynov’s pre-2015 analysis has showed Dendronotus robustus Verrill, 1870 that there are at least two species in the “D. Fig. 7 robustus complex”, namely D. robustus Verrill, Dendronotus robustus Verrill, 1870: 405; 1870 and D. velifer G.O. Sars, 1878. However, Lundin et al., 2017: 303, 306, 309–315, Figs 3, 4. this analysis was unfortunately not included Non auct., not Ekimova et al. (2015) (mix-up in the º»nal version of the publication by I.A. with D. velifer G.O. Sars, 1878). Ekimova (Ekimova et al., 2015). This detailed morphological and molecular analysis Extended diagnosis. Body wide. Six to seven has shown that D. robustus and D. velifer pairs of branched dorsolateral appendages. are well di¼ferentiated morphologically, 10 to 15 appendages of oral veil. Four to º»ve bathymetrically and according to the appendages of rhinophoral stalks. Lateral molecular analysis (Lundin et al., 2017). In papilla of rhinophoral sheaths absent. the previous literature these two species have Rhinophores with 7–14 lamellae. More than been mixed. 20 lip papillae. Basal colour light grayish to yellowish with scattered opaque white Dendronotus rufus O’Donoghue, 1921 and yellow spots. Dorsal processes of jaws Fig. 7 inclined posteriorly at approximately 47° Dendronotus rufus O’Donoghue, 1921:190– to the longitudinal axis of the jaw body and 192; Pl. 3, Figs 25–27; Pl. 4, Fig. 48; O’Donoghue, 0.55 of its length. Masticatory processes 1922:125; Robilliard, 1970: 456–460, Figs 4–6, apparently bear denticles (may possess ridge- 16–18, Pl. 63, Fig. 32. like structures). Radula with up to 31 rows of Behrens, 1980:76; Behrens & Hermosillo, teeth. Central tooth with high cusp and up to 2005:95; Lamb & Hanby, 2005: 262. 20 small distinct denticles without furrows. Dendronotus frondosus sensu Odhner, 1936 Up to nine lateral teeth with up to seven (part.) non Ascanius, 1774. denticles. Ampulla moderately narrow, with approximately three to four irregular loops. Extended diagnosis. Body relatively narrow. Bursa copulatrix moderate in size, rounded to Six to nine pairs of branched dorsolateral pear-shaped, with small seminal receptaculum appendages. Five appendages of oral veil. Five placed distally near the vaginal opening. appendages of rhinophoral stalks. Lateral Prostate discoid with about 10 alveolar glands. papilla of rhinophoral sheaths present. The vas deferens is moderate in length, penis Rhinophores with 19–24 lamellae. More massive, broad, slightly curved. Body length of than 30 lip papillae. Basal colour from red to veriº»ed specimens up to 50 mm. grayish white, white specimens commonly have remarkable di¼ferences in colour of Distribution. North Atlantic (both western and dorsolateral appendages (which can be brick eastern, Canada, Norway, Russia and USA), red to dark red). Dorsal processes of jaws very shallow waters only, in the Arctic, the inclined posteriorly at approximately 70°

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   to the longitudinal axis of the jaw body and veil. Three to seven appendages (posterior 0.45 of its length. Masticatory processess with and medial ones longest) of rhinophoral ridge-like structures and smaller denticles. stalks. Lateral papilla of rhinophoral sheaths Radula with up to 35 rows of teeth. Central absent. Rhinophores with 3–14 lamellae. tooth with up to 28 small denticles (usually 0–4 lip papillae. Basal colour opaque white, smaller) with ribs or shallow furrows, never solid orange, brownish, ranges from pale completely smooth. Up to 16 lateral teeth yellow to dark brown, scattered small yellow with up to seven denticles. Ampulla relatively spots, often with four thin subparallel lines narrow, long, with several windings and (two on each side) next to the dorsolateral loops. Bursa copulatrix large, oval. Seminal appendages. Dorsal processes of jaws receptaculum small, placed distally at a inclined posteriorly at approximately 65° moderately short distance from the vaginal to the longitudinal axis of the jaw body and opening. Prostate discoid with numerous 0.33 of its length. Masticatory processess alveolar glands (at least more than 50). The with ridge-like structures. Radula with up vas deferens is very long, penis long, narrow, to 72 rows of teeth. Central tooth with up looped several times. Body length up to to 10 small distinct denticles with furrows. 170 mm. Up to seven lateral teeth with up to seven denticles. Ampulla voluminous, folded. Bursa Distribution. NE Paciº»c, from southern Alaska copulatrix large, oval. Seminal receptaculum to Washington state, Canada and USA. small placed distally at a moderately short distance from the vaginal opening. Prostate Bathymetry. 7–38 m depth. discoid with up to 12 alveolar glands. The vas deferens is moderate in length, penis thin, Remarks. This is among the largest long, winded. Body length up to 40 mm. Dendronotus species. According to the radular formula it is similar to D. lacteus and molecular Distribution. North Eastern Paciº»c, Canada data conº»rm D. rufus as a sister species to D. and USA. lacteus (º»g. 1). In its biogeographical pattern it reveals quite narrow endemism, so far known Bathymetry. From intertidal to 20 m depth. only from southern Alaska to Washington state. Remarks. Dendronotus subramosus is unique among the clades of narrow-bodied Dendronotus subramosus MacFarland, 1966 species since it has no lateral papillae Fig. 7 on the rhinophoral sheath. However, Dendronotus subramosus MacFarland, phylogenetically D. subramosus is more 1966: 265–270, Pl. 40, Fig. 3; Pl. 46, Figs 5–8; Pl. closely related to the externally similar- 47, Figs 3–7; Pl. 49, Figs 1–3; Pl. 50, Fig. 2; Pl. 52, looking deep sea species D. jamsteci Figs 1, 2; Robilliard, 1970: 462–466, Figs 3, 4, 6, (with the presence of rhinophoral sheath 19–21, Pl. 64, Fig. 33; Behrens, 1980:76; Behrens papillae), which was recently described & Hermosillo, 2005:95; McDonald, 1983: 177. o¼f Paciº»c Honshu (Japan) and also to the externally di¼ferent-looking NE Paciº»c Extended diagnosis. Body narrow. Three shallow water species D. albus (º»gs 1, 2). Such to six pairs of branched dorsolateral a morphological and molecular evolutionary appendages. Four to six appendages of oral mosaic is remarkable and contribute for the

 $        # $# !"%   %   ./-  |   . fulº»lling of the term “mutlilevel º»ne-scale Dendronotus venustus MacFarland, 1966 organismal diversity”. Fig. 7 Dendronotus venustus MacFarland, 1966: 271– Dendronotus velifer G.O. Sars, 1878 275, Pl. 40, Fig. 2; Pl. 46, Figs 9–12; Pl 47, Figs Fig. 7 1, 2; Pl. 49, Fig. 6; Pl. 50, Fig. 3; Pl. 52; Figs 3–6; Dendronotus velifer G.O. Sars, 1878: 238–239, Stout et al. 2010: 7. Pl. 15, Fig. 4, Pl. 28, Fig. 2; Lundin et al., 2017: D. frondosus sensu Robilliard, 1970 and auc- 303–309, 310–318, Figs 1, 2, 4. torum non Ascanius, 1774. Non D. velifer auct. (mix-up with D. robustus and D. bathyvela). Diagnosis. Body narrow. Four to eight pairs of branched dorsolateral appendages. Four Extended diagnosis. Body wide. Four to six to eight appendages of oral veil. Four to six pairs of branched dorsolateral appendages. appendages (posterior and medial ones Six to twelve main appendages on oral veil. longest) of rhinophoral stalks. Lateral papilla Four to º»ve appendages on rhinophoral stalks. of rhinophoral sheaths present. Rhinophores Lateral papilla of rhinophoral sheaths absent with 8–14 lamellae. Lip papillae absent or or very small. Rhinophores with 12–15 lamellae. up to eight. Colour forms include white 30–35 lip papillae. Basal colour commonly specimens with yellow spots and brownish light-to bright red with scattered white dots. or reddish specimens with yellow dots and/or Dorsal processes of jaws inclined posteriorly tubercles or large amount of white pigment, at approximately 65° to the longitudinal but with no white-tipped appendages. Dorsal axis of the jaw body and 0.55 of its length. processes of jaws inclined posteriorly at Masticatory processes with denticles. Radula approximately 55–60° to the longitudinal with up to 36 rows of teeth. Central tooth axis of the jaw body and 0.41 of its length. with low to moderate cusp and up to 26 small Masticatory processess with ridge-like distinct denticles without furrows. Up to 15 structures and smaller denticles. Radula with lateral teeth with which is smooth or with up up to 48 rows of teeth. Central tooth with up to four rudimentary denticles. Ampulla two- to 12 small distinct denticles with furrows. Up looped. Bursa copulatrix large, rounded, with to eight lateral teeth with up to º»ve denticles. small seminal receptaculum placed distally Ampulla voluminous, folded. Bursa copulatrix at some distance from the vaginal opening. large, pear-shaped. Seminal receptaculum Prostate discoid with more than 50 alveolar small placed distally at considerable distance glands. The vas deferens is moderate in length, from the vaginal opening. Prostate discoid penis thin, long. Body length up to 90 mm. with few (5–12) alveolar glands. The vas deferens is long, possibly also partly prostatic Distribution. North Atlantic (southern limit – since thickened on considerable length and Gullmarºìord, Sweden) and Arctic (at least to then distally narrowed, penis thin, long. Body the Laptev Sea), Norway and Russia. length up to 42 mm.

Bathymetry. Commonly found below 50 m Distribution. NE Paciº»c, Canada and USA. depth, in the Arctic not shallower than 10 m. Bathymetry. Common in shallow waters, at Remarks. See Remarks for D. robustus. depths of circa 5–20 m.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Remarks. Despite MacFarland’s (1966) Five to seven appendages of oral veil. Circa detailed description of D. venustus, it was º»ve appendages (posterior ones the longest) almost universally considered a synonym of rhinophoral stalks. Lateral papilla of of D. frondosus (Robilliard, 1970; McDonald, rhinophoral sheaths present. Rhinophores 1980; Behrens & Hermosillo, 2005) until with 10–11 lamellae. Lip papillae 20–30. Basal molecular data showed that it is distinct from colour bright red to reddish-brownish with D. frondosus (Stout et al., 2010; Korshunova et thin white broken lines between dorsolateral al., 2019b). Robilliard (1970) did not doubt the appendages, also scattered opaque white dots synonymy of D. venustus, although he rarely and speckles on dorsal and lateral sides, and observed more than 12 of the prostatic alveolar on various appendages. Dorsal processes of glands in the NE Paciº»c specimens, whereas jaws inclined posteriorly at approximately in true D. frondosus their number is usually up 70° to the longitudinal axis of the jaw body to at least 30. There are also minor di¼ferences and 0.5 of its length. Masticatory processess in the radular patterns and some details of with ridge-like structures. Radula with up to colouration between D. frondosus and D. 36 rows of teeth. Central tooth completely venustus (e.g., common presence of white devoid of denticles (except for the anterior- forms with small yellow spots in D. venustus). most juvenile radula in some specimens), There is a possibility of the presence of hidden only sometimes faint traces of furrows diversity within D. venustus. However, a occur. Up to 12 lateral teeth with up to seven specimen from Alaska included in our present denticles. Ampulla moderately narrow, study (º»g. 7) matches the original description sligthly bent. Bursa copulatrix very large, of D. venustus very well by external and internal oval. Seminal receptaculum small placed features (MacFarland, 1966: 271–275, pl. 40 º»g. distally at considerable distance from the 2, pl. 46, º»g. 9, pl. 50 º»g. 3). These characters vaginal opening. Prostate discoid with circa 25 include the presence of yellow-greenish spots alveolar glands. The vas deferens is moderate on a greyish to whitish body, a speciº»cally in length, penis conical, curved. Body length broadened, somewhat cheliform º»rst lateral up to 30 mm. tooth, and a long, broadened, apparently partly prostatic vas deferens in addition to a small Distribution. NW Paciº»c, so far recorded discoid alveolar prostate. We therefore surely from Kamchatka, Russia, southern part of refer to this specimen as a true D. venustus in Hokkaido and Honshu, Japan (the Sea of our present phylogenetic analysis (º»g. 1). Japan side).

Dendronotus zakuro Martynov, Fujiwara, Bathymetry. Shallow waters, at depths of circa Tsuchida, R. Nakano, N. Sanamyan, 7–20 m. K. Sanamyan, Fletcher & Korshunova, 2020 Fig. 7 Remarks. This species has recently been Dendronotus zakuro Martynov, Fujiwara, described in detail (Martynov et al., 2020a). Tsuchida, Nakano, Sanamyan, Sanamyan, Fletcher, Korshunova, 2020a: 505–507, Figs 3, Genus Cabangus gen. nov. 5B. ZooBank: http://urn:lsid:zoobank.org:act: 3B779CBF-A310-4B47-85B1-488104FAD541 Extended diagnosis. Body narrow. Six to seven Type species. Dendronotus regius Pola & pairs of branched dorsolateral appendages. Stout, 2008

 $        # $# !"%   %   ./-  |   .

Etymology. From the Indonesian word denticle-bearing part of the central teeth is not “cabang” meaning “branch” in reference to separated from the lateral sides by a distinct this genus as “dendronotids of the tropics” shoulder. Thus the central teeth have a squarish and to respect the great contribution of base and a triangular top. This makes the the Indonesian fauna to global marine radula of both species more similar (though biodiversity (e.g., Hoeksema, 2007). not identical) to that of the family Bornellidae, whose species are also partly similar in external Diagnosis. Dorsolateral appendages with morphology. However, details of the stomach distinct tertiary branches. Oral veil with and the reproductive systems are di¼ferent. branched appendages. Radula with at least Therefore, taking molecular and morphological up to nine lateral teeth in adult specimens. evidence into account, we propose here the Central teeth with cusp integrated within new genus Cabangus gen. nov. with as type lateral denticles. Denticle-bearing part of species Dendronotus regius Pola & Stout, 2008. central teeth not separated from lateral sides, The holotype of the only other tropical species distinct shoulders absent, thus central teeth described so far, D. noahi, is likely a juvenile triangular in outline. Prostate represented by since it possesses only four rows of lateral teeth a thickened structure without evident alveolar and immature reproductive system. Although glands and disk. Copulative organ conical. this species is now included in the genus Cabangus gen. nov., this needs further study. Remarks. According to the present molecular Externally, Cabangus spp. are di¼ferent from the phylogenetic analysis, the tropical Dendronotus common external appearance of Dendronotus regius Pola & Stout, 2008 comes as a most basal spp. by a combination of a very narrow clade, distinct from all presently known species body and short dorsolateral appendages. By of the genera Dendronotus and Pseudobornella. these characters, Cabangus gen. nov. is also Morphologically, D. regius di¼fers from any somewhat similar to the family Bornellidae, known species of the genus by the presence which is phylogenetically distant from the of a homogenous, º»ne prostate, which is Dendronotidae. Cabangus gen. nov. represents externally smooth and does not show any a basalmost clade within the Dendronotidae alveoles. Pola and Stout (2008: Fig. 4) evidently and hence may retain some features of the did not indicate any alveols in the prostate common ancestors with phylogenetically more of their D. regius, but in the text they (Pola & distantly related families. Cabangus gen. nov. Stout, 2008: 48) mentioned a “large prostate is also well supported by biogeographic data with the proximal limit being marked by a because the majority of its species inhabits closely set ring of alveolar glands”. To clear this Arctic and temperate waters, whereas Cabangus contradiction between this º»gure and the text gen. nov. represents a distinct tropical lineage. in the original description, we checked several adult specimens of D. regius and conº»rm Cabangus noahi (Pola & Stout, 2008) comb. that there are no externally or internally nov. conspicous alveols, otherwise so characteristic Dendronotus noahi Pola & Stout, 2008: 55–63, for a majority of true Dendronotus species. º»gs 6A, B. There is another striking di¼ference between both strictly tropical species of Dendronotus Diagnosis (original description). Body very (D. regius plus D. noahi) and any other narrow. Four pairs of branched dorsolateral known species of this genus, namely that the appendages. Six appendages of oral veil. Four

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   appendages (posterior longest) of rhinophoral seven denticles. Ampulla voluminous, folded. stalks. Small lateral papilla of rhinophoral Bursa copulatrix large, pear-shaped to oval. sheaths possibly present. Rhinophores with Seminal receptaculum small placed distally nine lamellae. Lip papillae not indicated in the at a short distance from the vaginal opening original description. Colour semitransparent (in the original description entangled duct white with brownish branches of digestive from ampulla was incorrectly identiº»ed gland. Masticatory processes with denticles. as “receptaculum seminis”). Prostate non Radula with up to 18 rows of teeth. Central discoid, thickened, without evident alveolar tooth quite narrow, with shallow furrows glands. The vas deferens is moderate in length, and with up to 15 distinct denticles. Up to penis tapered. Body length up to 15 mm. four lateral teeth with up to eight denticles. Reproductive system reported as immature in Distribution. Tropical Indo-west Paciº»c. the º»rst description. Body length up to 4 mm (juvenile specimen). Bathymetry. Up to 45 m deep.

Distribution. Papua New Guinea, north coast, Remarks. See under Cabangus gen. nov. outer barrier reef, Bagabag Island, Bismarck Sea. Genus Pseudobornella Baba, 1932 Type species. P. orientalis Baba, 1932 (º»g. 2) Bathymetry. 30.5 m depth. Diagnosis. Dorsolateral appendages without Remarks. Molecular data unavailable, as well distinct (or with few very short) tertiary as additional specimens. Species is in need of branches. One of the rhinophoral stalk further study. appendages from both sides is extremely long, usually exceeding body length. Oral veil with Cabangus regius (Pola & Stout, 2008) comb. long simple unbranched appendages. Radula nov. with very small number of lateral teeth Fig. 2 (presently known to be no more than two rows Dendronotus regius Pola & Stout, 2008: 46– of lateral teeth). Central teeth with protruding 54, Figs 1–5. cusp distinctly separate from lateral denticles. Prostate represents by narrow tube, without Extended diagnosis. Body very narrow. alveols and disk. Copulative organ partly Three to four pairs of branched dorsolateral ÐÑattened and with a widened subcircular appendages. Circa six appendages of oral apical part. veil. Four to º»ve appendages (posterior ones longest) of rhinophoral stalks. Rhinophores Remarks. The genus Pseudobornella di¼fers from with 10–11 lamellae. Lateral papilla of both Dendronotus and Cabangus gen. nov. by rhinophoral sheaths present. Lip papillae a unique combination of external and internal possibly absent. Basal colour white or yellow, characters. These characters include a very long with blue, brown and reddish patches. appendage of the rhinophoral sheath, absence Masticatory processes with denticles. Radula of distinct tertiary branches of the dorsolateral with up to 36 rows of teeth. Central tooth appendages, resulting in the general appearance with up to 20 small distinct denticles with of dorsolateral appendages of Pseudobornella as furrows. Up to nine lateral teeth with up to ctenidium-like. Instead, all species of the genus

 $        # $# !"%   %   ./-  |   .

Dendronotus possess well-deº»ned tertiary coupled with the molecular data (including branches on the dorsolateral appendages, and no presenting for the º»rst time data on Û gene species has such a exceedingly long appendages for the type species Pseudobornella orientalis, of the rhinophoral sheaths. There is a deep-sea º»gs 1, 2) show that the genus Pseudobornella species, Dendronotus claugei (see above), for represents an early o¼fshoot of the family which unbranched dorsolateral appendages are Dendronotidae. This evolution was fuelled reported, but this species is only known from by the paedomorphic reduction of the lateral a single specimen and needs further studies. teeth and juvenilization of the central teeth According to the present phylogeny D. claugei (for criteria of paedomorphosis (Korshunova may represent a separate genus, but more data et al., 2020a; Martynov et al., 2020b). A pre- are needed to support this. Furthermore, so far vious morphological cladistic study (Pola et no single species of the genus Dendronotus and al., 2009) placed the genus Pseudobornella Cabangus gen. nov. have smooth unbranched outside the family Dendronotidae, thus appendages in the oral veil, whereas numerous highligthing the morphological peculiarities reported Pseudobornella specimens invariably of this genus as inconsistent with those of show strong unbranched appendages of the other dendronotid genera. However, a solely oral veil. Ultimately, the radular pattern of molecular study suggested to synonymize the genus Pseudobornella di¼fers from that in Pseudobornella with Dendronotus (Pola & the adult stage of species of Dendronotus and Gosliner, 2010). We conº»rm here the validity Cabangus gen. nov. by a very small number of of the genus Pseudobornella Baba, 1932 using lateral teeth (so far reported no more than two). integrative evidences. Specimens of the type Adult specimens of the genera Dendronotus species Pseudobornella orientalis from the Sea and Cabangus gen. nov. possess at least eight of Japan (º»g. 2) in the present study matched lateral teeth, and only very rarely in poorly morphologically well with those in the orig- known deep-sea lineages, the number of the inal description of P. orientalis (Baba, 1932), lateral teeth rows can reach six in number. The including the characteristic small choco- tropical C. noahi was reported to have only four late-brown spots and yellow lines on the dor- rows of lateral teeth, but only a single juvenile sal side, and the shape of the radular teeth. specimen is known (Pola & Stout, 2008). All investigated species of the genus Dendronotus Species composition. This presently mono- showed the presence of 1–3 rows of lateral teeth speciº»c genus only contains P. orientalis Baba, during the early stages of their ontogeny (e.g., 1932. Martynov et al., 2020a), while only in the genus Pseudobornella such feature became apparent in adults stage. Furthermore, the shape of the Discussion central teeth of Pseudobornella is also di¼ferent from that in any adult Dendronotus and General taxonomic and biogeographic Cabangus gen. nov. species by the presence of a overview of the family Dendronotidae: strongly protruding central cusp on the central a revolution of ne-scale dening of species teeth, which is distinctly separated from the The family Dendronotidae and its type genus lateral denticles. Dendronotus represent an emerging model A Ø image of the radula of P. orientalis from for studying speciation, and ultimately for the Sea of Japan was presented in Martynov investigating general problems related to et al. (2015a: 60, º»g. 5G). Radular characters the “species concept”. The total number of

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   species belonging to the genus Dendronotus Dendronotus europaeus inhabits the north- in restricted sense is 27 (a review of all the erneastern Atlantic in European waters only. previously described and species consid- Dendronotus arcticus is reported only from ered valid with diagnoses is presented in the the true Arctic, e.g., Kara and Laptev seas. Synopsis section). Though some new spe- Dendronotus lacteus occurs in northeastern cies were recognized using morphological Atlantic and Arctic with the Laptev Sea as methods (e.g., MacFarland, 1966; Robilliard, easternmost limit. Dendronotus yrjargul sp. 1970, 1972), the species composition of the nov. inhabits mostly the Arctic (with the Kara genus Dendronotus remained quite conserv- Sea as easternmost limit) and the neighbour- ative (Thompson & Brown, 1984; Roginskaya ing northern European regions with middle 1987, 1997). The genus Dendronotus was Norway as southernmost limit. Dendronotus represented by just about 10 recognised nordenskioeldi sp. nov. is so far only reported shallow-water species worldwide (plus one from the Arctic Laptev sea. Dendronotus deep-sea and two or three species of unclear kalikal is distributed in the northwestern status) until a considerable amount of hid- Paciº»c until the Bering Strait. Dendronotus den diversity was discovered recently in the kamchaticus is limited to the Northern Paciº»c shallow waters of both the North Paciº»c and only, with scattered records from the north- the North Atlantic (e.g., Korshunova et al., western and northeastern parts. Dendronotus 2017a; Martynov, 2020a). zakuro is reported from the northwestern The conservative approach in the tax- Paciº»c only, ranging from middle Honshu onomy of the family Dendronotidae relied to Kamchatka Peninsula. Dendronotus pri- considerably on Mayr’s (1942, 1969) polytypic morjensis is restricted merely to the Sea of species concept and allowed the presence Japan and neighbouring localities in Japan. of broadly deº»ned “species” distributed over Dendronotus venustus is restricted to the very large geographic distances. For example, northeastern Paciº»c. This considerable until about the year 2000 only D. frondosus regionalisation in a majority of the species is was recognised among the North Atlantic remarkable since all of them have planctonic narrow-bodied Dendronotus spp. (Thompson veliger larvae and may potentially disperse & Brown, 1984). The wide geographic range of very widely. But in reality instead of a single D. frondosus at that time also encompassed “inº»nitely variable” species around whole the entire North Paciº»c up to subtropical Eurasia we have an enormously complicated regions of the middle of Honshu (Japan) and radiation of evidently separate species. From China (e.g., Robilliard, 1970; Baba, 1993; Lin these species, so far only D. lacteus shows a et al., 1986). Later this has showed not to be maximum range from the northern European the case, and the supposedly super-polytypic seas to the Arctic. However, it also shows a and pan-geographic “species” D. frondosus considerable morphological variation com- has been split into at least 11 narrow-deº»ned pared to other species, which may lead to this species (see below and º»gs 1, 7). Each of these species becoming further subdivided. A very species occupies a particular geographic broad geographic range (the North Atlantic, region. The natural range of true D. frondo- Arctic and North Paciº»c) was previously also sus sensu stricto is limited solely to the east- indicated for Dendronotus dalli (Robilliard, ern and western parts of the North Atlantic 1970; Roginskaya, 1987). Mutilevel data with the subarctic Barents Sea as easternmost for the putative North Atlantic and Arctic limit, while it is absent in the Arctic region. “D. dalli” show instead that it represents a

 $        # $# !"%   %   ./-  |   . separate species D. elegans (as “D. niveus”). Notably, Robilliard (1970) noted di¼ferences in Dendronotus elegans is a predominatly Arctic the reproductive system between European species, which reaches cold water masses at D. frondosus and his NE Paciº»c specimens of the Canadian and US North Atlantic coasts. D. venustus that he at that time identiº»ed as Dendronotus dalli instead is a predominantly “D. frondosus”. At the same time, apparently North Paciº»c species, which reach Arctic key reproductive features (e.g., number of seas into the Bering Strait and neighbouring prostatic alveols and shape of ampulla) can waters, but it is absent in the subarctic North vary signiº»cantly, as in the cases of D. albus or Atlantic regions. D. elegans. The separate characters should be Importantly, many of these species could therefore used with considerable care to dis- be separated by a careful application of mor- tinguish particular species. phological characters without aid of molec- All these cases clearly conº»rm that not ular data. The possibility of morphological only molecular data, per se, changed the separation is evident in the case of D. venustus modern landscape of biodiversity research. from the NE Paciº»c. Dendronotus venustus was The fundamental changes have occurred described in considerable detail and already in the dominated paradigm of the species became delineated from the resembling D. description. A major modern feature of this frondosus in the mid-20th century (published is the general denial of huge polytypic and in MacFarland, 1966) without use of molecular pan-geographic species in favour of º»ne-scale techniques and scanning electron microscopy interspeciº»c morphological di¼ferences in (Ø). Ø has been available to taxonomists combination with limited geographic ranges. since the 1970s and its use would enable dif- Within restricted “narrow species” a polymor- ferentiation of some of the Dendronotus spe- phism (a basis for the polytypic concept) is cies that were described in the 21st Century. often manifested in the occurrence of parallel However, this did not happen because all these morphs, which may obscure species diversity characters could be considered as too subtle (see e.g., Korshunova et al., 2020b). Though from the perspective of the past dominant Mayr was among the major developers of the paradigm. For instance, Thompson and Brown population-based, variable, “non-typological” (1984) published an Ø image of radula of a species concept, he also admitted the possi- specimen identiº»ed as D. frondosus but which ble existence of a hardly detectable diversity actually belongs to the recently described D. of “sibling species” (Mayr, 1942; Yoder et al., europaeus (Korshunova et al., 2017b). Even 2005). Despite this, the gravity of “a morpho- such morphologically well distinguished logically super-variable species concept” was species as D. albopunctatus (Robilliard, 1972) clearly unfavourable for seeking of º»ne-scale was commented as very similar to D. robus- di¼ferences. This was one of the reasons why tus (see Rudman, 2007). While data on the researchers for a long time omitted evidence multilevel diversity within Dendronotidae for morphological di¼ferences in the “D. fron- are currently accumulating, more and º»ner dosus megacomplex”. The polytypic concept cases will be discovered in the future among became popular more than º»ve decades already restricted “species”. Nevertheless, a ago and at that time Mayr (1942) suggested majority of the recently separated species can a potential solution for how to unfold the either be di¼ferentiated by Ø radula data or apparent “chaotic species” of “typologists” by a combination of the habitus data, radular into a polymorphic “biological” species. data and features of the reproductive system. However this concept can now be considered

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   as rather an obstacle than an aid in the dis- not yet described diversity within the genus covery of real patterns of biodiversity. As a Dendronotus was discovered in the temperate clear evidence for that, until very recently waters of Southern Hemisphere (e.g., Burn, only drastically di¼ferent species within the 2015). genus Dendronotus have been recognised Most of the previously recognised without doubts (Robilliard, 1970; McDonald, Dendronotus species (seven out of nine, see 1983), despite uncertanties in their delimita- Robilliard, 1970, 1972) occur in the NE Paciº»c, tion (Robilliard, 1975). An especially remarka- while in the North Atlantic and NW Paciº»c ble example is the notorious problem of how only one or two species were commonly rec- to distinguish “D. albus” from “D. diversicolor”, ognized. This caused a distorted picture of a which has persisted since Robilliard (1970). biogeographic asymmetry in the Dendronotus This problem was in a profound confusion of diversity between the North Atlantic, north- the morphological characters rather than in western Paciº»c and northeastern Paciº»c. One the absence of molecular data and was solved of the important results of the º»ne-scale spe- only recently by the distinction of D. robil- cies deº»nition is a considerable reduction liardi (Korshunova et al., 2016a). of the “biogeographic asymmetry” among The majority of the described species of the Dendronotus spp. between NE Paciº»c, NW genus Dendronotus inhabit cold and temper- Paciº»c, Arctic and North Atlantic localities. ate regions of the Northern Hemisphere and The resulting numbers of shallow-water and their diversity clearly declines towards trop- shelf species in such regions, as the North ical localities. Pola and Stout (2008) assigned Atlantic and the NW Paciº»c, are impressive. two tropical dendronotids º»rst to the genus I.e., in the North Atlantic and western subarc- Dendronotus. However, the morphology of the tic regions (Barents Sea as limit) there are cur- male part of the reproductive system of the rently seven species, in the true Arctic (Kara, these tropical representatives is di¼ferent from Laptev and East Siberian seas) there are also those in the genus Dendronotus, and available seven species so far described and recorded, molecular data for “D. regius” show it to be a sep- while in the NW Paciº»c (Bering Strait as north- arate with a more basal placement compared ernmost limit) there are eigth species and in to any other dendronotid (º»gs 1, 2). Therefore NE Paciº»c there are 10 species. In all regions “D. regius” has become the type species of the the number of species will likely increase fur- new genus Cabangus gen. nov., which encom- ther. This analysis clearly suggests that not passes exclusively tropical dendronotids. This only geographical proximity to the rich warm corroborates well with the new general agenda water fauna (as in the case of NE Paciº»c) is of the multilevel organismal diversity, which important, but also that ecological and other proposes narrowly deº»ned taxa not only on a factors (including the history of faunal for- species level, but also at the genus and family mation) have contributed substantially to levels (Korshunova et al., 2017a, b, c; 2019a). an active speciation. The presence of at least This prevents to produce non-diagnosable seven species of Dendronotus in the Artic as huge taxa that not only lack support from mor- one of the world’s coldest regions is a clear evi- phology but also from various other data, such dence for this. All these data and implications as biogeography. The taxonomic diversity of have contributed to rapid and revolutionary the family Dendronotidae undergoes a process advancements in the taxonomy of the family of understanding and more taxa in this family Dendronotidae (Thollesson, 1998; Stout et al., are expected to be discovered. For example, a 2010; Martynov et al. 2015a, c, 2020a; Ekimova

 $        # $# !"%   %   ./-  |   . et al. 2015; Korshunova et al. 2016a, b, 2017a, genus Dendronotus implies the presence of 2019b; Lundin et al., 2017; present study). many taxa that are incomparably more close to D. frondosus as the type species of the genus Multilevel Dendronotidae diversity and (e.g., D. primorjensis, D. venustus) than to the di ferent degrees of evolutionary separation much di¼ferent wide-bodied D. robustus (and The majority of dendronotids are putatively its closely related species, e.g., D. velifer, D. united under the genus Dendronotus, which bathyvela). Therefore, although a direct com- represents a morphologically heterogenous parison between D. robustus sensus strictus assemblage of several distinct clades with a and D. frondosus sensus strictus is possible, signiº»cant molecular divergence (see º»gs 1, 2; this is not taxonomically meaningful any- Synopsis; Appendix, table A2). The case of the more considering the presence of such hardly family Dendronotidae is therefore particularly distinguishable complexes. The similarities illustrative of multilevel di¼ferences among between the “Dendronotus robustus complex” apparently taxonomically equally recognized and the “D. frondosus megacomplex” are thus “species”. These apparently natural units are similarities at genus or higher level. This needs not equal by their morphological and molecu- to be adressed in a further study. lar properties, and broader, ontogenetic prop- The common formulation of a procedure erties. Hence, the term “species” is not equal to º»nd di¼ferences within such multilevel over large phylogenetic distances but also organismal diversity, and hence the degree of among relatively small levels, such as genus separation between di¼ferent groups of bio- and family. To make taxonomy consistent logical organisms is a “species delimitation” with the multilevel diversity among the fam- despite the species concept itself having no ily Dendronotidae, we separate a new genus universal fundamental agreement even in Cabangus gen. nov. and conº»rm a separate recent publications (Stanton et al., 2019). In status of the genus Pseudobornella. the present study we found a remarkable case Notably, even among the “core species” when the mean Û di¼ference of a group is that is still constitute the genus Dendronotus slightly above 2% (normally indicative of (º»g. 1) our present study demonstrates various a single species) but when morphological, degrees of separation from species in a state of ontogenetic and biogeographic evidence being morphologically well distinguished but are examined the two Dendronotus groups molecularly low delineated (the D. yrjargul deserve to be taxonomically marked as dis- sp. nov. – D. kalikal pair) to morphologically tinct species, namely D. kalikal and D. yrjar- subtle but molecularly well distinguished gul sp. nov. (º»gs 1–4, 6). At the same time species (D. frondosus–D. primorjensis, D. dal- D. nordenskioeldi sp. nov., despite its habitus li–D. elegans, D. lacteus–D. europaeus and similarity with D. lacteus, di¼fers substantially D. nordenskioeldi sp. nov.) (º»gs. 1–8). We can from that species by its central teeth with unambiguously present distinguishing char- very weak denticles devoid of rib-like struc- acters only when these apparent groups tures (º»g. 5) and intriguingly, (º»g. 1) it demon- currently known as species really belong to strates signiº»cant molecular distance from su¼º»ciently distantly related subclades, as in all know species of the genus Dendronotus easily recognizable species such as D. iris or (Appendix, table A2). Despite the lack of D. subramosus, or D. frondosus and D. robus- consensus about the meaning of “species” we tus (º»gs 1, 7). A very complicated mutilevel can observe some “groups of genetically and morphological and molecular diversity of the ecologically similar individuals” (e.g., Shapiro

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   et al., 2016) in nature, even though there are of live specimens. Our warm gratitude goes to no clear criteria about how to separate one Karin Fletcher for improvement of English in similar group from another similar group. an earlier version of the manuscript and for A conclusion was reached that the “popula- providing us with the wonderful pictures of tion-to-species” level represents a continuum D. rufus and D. venustus. Electron Microscopy (Mallet et al., 2009; Coates et al., 2018). Using Laboratory MSU is thanked for support these numerous available data on di¼ferent with electron microscopy. Reviewers are groups and our present data, we therefore dis- thanked for providing comments. The study agree with the most recent suggestion to save was supported by the Norwegian Taxonomy “species” as a universal concept using a “weak Initiative project #sneglebuss Barents Sea (19- realism” approach (Reydon & Kunz, 2019). 18_70184240). The work of AM was supported Because even in its weaker form, “species” as by the research project of MSU Zoological a reality still implies the universal application Museum (AAAA-A16-116021660077-3). The of this concept despite that it encompasses work of TK was conducted under the IDB RAS groups that have di¼ferent degrees of separa- Government basic research program in 2020 tion, and hence di¼ferent morphological and № 0108-2019-0002. molecular properties. The big discrepance between “species” or other taxon as a sys- tematic unit and “species/taxon” as a natu- References ral entity still persists (e.g., Ereshefsky, 1998; Zachos, 2018). The recently proposed con- Agarwal, M. (2017) First record of Dendronotus cept of the multilevel organismal diversity orientalis (Baba, 1932) (Nudibranchia: Den- (Korshunova et al., 2017b, 2019a) is helping dronotidae) in the temperate Eastern Paciº»c. to reduce this discrepance and make a tran- BioInvasions Rec., 6, 135–138. sit from the previous paradigm of “polytypic Alder, J. & Hancock, A. (1842) Descriptions of species”, by placing the organismal diversity several new species of nudibranchous mollusca in a broad ontogenetic framework, as a º»ne- found on the coast of Northumberland. Ann. scale complex of the morphological, genetic Mag. Nat. Hist., 9, 31–36. and epigenetic patterns and processes. Ascanius, P. (1774) Beskrivelse over en Norske sneppe og et sodyr. Det Kongelige Norske Videnskabelige Selskabs Skrifter Trondhejm, Acknowledgements 5, 153–158. Baba, K. (1932) Pseudobornella orientalis, nov. gen. Olga Zimina is warmly thanked for pro- et sp. from Japan. Annot. Zoolog. Japon., 13, 4, viding specimens from Arctic. The team of 369–376. Gulen Dive Center (Christian Skauge, Ørjan Baba, K. (1949) Opisthobranchia of Sagami Bay Sandnes, Monica Bakkeli, and Guido Schmitz) collected by His Majesty The Emperor of Japan. are generously thanked for their help during Iwanami Shoten, Tokyo. º»eldwork in Norway, as these Norwegian spec- Baba, K. (1993) A northern species of Dendronotus imens were used for comparative purposes (Mollusca: Nudibranchia: Dendronotidae) from in this study. We are thankful to Yoshihiro Sado Island, Sea of Japan. Rep. Sado Mar. Biol. Fujiwara, Kazunori Hasegawa, Doug Miller, Stat., 23, 29–33. Karen Sanamyan, Nadezhda Sanamyan, and Bakken, T., Hårsaker, K. & Daverdin, M. (2020) Olga Zimina for providing some photographs Marine invertebrate collection NTNU University

 $        # $# !"%   %   ./-  |   .

Museum. Version 1.414. NTNU University Museum. Colgan, N. (1914) The opisthobranch fauna of the Occurrence dataset https://doi.org/10.15468/ shores and shallow waters of County Dublin. Ir. ddbs14 accessed via GBIF.org on 2020-02-25. Nat. J., 23 (8/9), 161–204. Becher, E. (1886) Mollusken von Jan Mayen, Cooper, J. (1863) On new or rare Mollusca gesammelt von Dr. F. Fischer, Arzt der inhabiting the coast of California. No. II. Proc. österreichishcen Expedition auf Jan Mayen. Calif. Acad. Nat. Sci., 3, 56–60. Die internationale Polarforschung, 1882/83, 3, Couthouy, J. (1838) Descriptions of new species 1–16. of Mollusca and shells, and remarks on several Behrens, D. & Hermosillo, A. (2005) Eastern Paci˜™c Polypi found in Massachusetts Bay. Boston J. nudibranchs, a guide to the opisthobranchs from Nat. Hist., 2, 53–111. Alaska to Central America. Sea Challengers, Crossman, C., Taylor, E., & Barrett-Lennard, L. Monterey. (2016) Hybridization in the Cetacea: widespread Behrens, D. (1980) Paci˜™c Coast Nudibranchs, a occurrence and associated morphological, Guide to the Opisthobranchs of the Northeastern behavioral, and ecological factors. Ecol. Evol., 6, Paci˜™c. Sea Challengers, Los Osos, CA. 1293–1303. Bergh, R. (1863) Campaspe pusilla, en ny Darwin, C. (1859) On the Origin of Species by Means Slægstform af Dendronotidernes Gruppe, of Natural Selection. John Murray, London. samt Bemærkninger om Dotidernes Familie. de Queiroz, K. (2006) Ernst Mayr and the modern Naturhist. Tidsskr. Stiftet af Henrik Kroyer, 1, concept of species. Proc. Nat. Acad. Sci. USA, 471–483. 102, 6600–6607. Bergh, R. (1879) On the nudibranchiate gasteropod de Queiroz, K. (2007) Species concepts and mollusca of the North Paciº»c Ocean, with species delimitation. Systematic Biology, 56, special reference to those of Alaska. Proc. Acad. 879–886. Nat. Sci. Philadelphia, 31, 71–132. Ekimova, I., Korshunova, T., Schepetov, D., Bergh, R. (1886) Die Nudibranchien. Gesammelt Neretina, T., Sanamyan, N. & Martynov, A. wärhrend der fahrten des “Willem-Barents” in (2015) Integrative systematics of northern and das Nördliche Eismeer. Bijdr. Dierk., 13, 1–37, pls. Arctic nudibranchs of the genus Dendronotus 1–3. (Mollusca, ), with descriptions Bergh, R. (1904) Malacologische Untersuchungen. of three new species. Zool. J. Linn. Soc., 173, Reisen im Archipel der Philippinen von Dr. Carl 841–886. Gottfried Semper, 9, 1–56. Ekimova, I., Schepetov, D., Chichvarkhina, O. & Bickford, D., Lohman, D.J., Sodhi, N.S., Ng, Chichvarkhin, A. (2016) Nudibranch molluscs P.K.L., Meier, R., Winker, K. & Das, I. (2006) of the genus Dendronotus Alder et Hancock, Cryptic species as a window on diversity and 1845 (: Dendronotina) from conservation. Trends Ecol. Evol, 22, 148–155. Northwestern Sea of Japan with description of a Bleakney, J. (1996) Sea slugs of Atlantic Canada and new species. Invertebr. Zool., 13, 15–42. the Gulf of Maine. Nimbus Publishing & The Eldredge, N. & Gould, S.J. (1972) Punctuated Nova Scotia Museum, Halifax. equilibria: an alternative to phyletic gradualism. Burn, R. (2015) Nudibranchs and Related Molluscs. In Schopf T.J.M. (Ed.) Models in Paleobiology, Museum Victoria, Melbourne. pp. 82–115. Freeman Cooper, San Francisco. Coates, D., Byrne, M. & Moritz, C. (2018) Genetic Eliot, C. (1910) A Monograph of the British diversity and conservation units: dealing with Nudibranchiate Mollusca: with Figures of the the species-population continuum in the age of Species. pt.  (supplementary). Ray Society, genomics. Front. Ecol. Evol., 6, 165. London.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

Ereshefsky, M. (1998) Species pluralism and anti- Katoh, K., Misawa, K., Kuma, K. & Miyata, T. (2002) realism. Philos. Sci., 65, 103–120. MAFFT: a novel method for rapid multiple Evertsen, J. & Bakken, T. (2005) Nudibranch sequence alignment based on fast Fourier diversity (Gastropoda, Heterobranchia) along transform. Nucleic Acids Res., 30, 3059–3066. the coast of Norway. Fauna Norvegica, 25, 1–37. Kijewski, T., Zbawicka, M., Väinölä, R. & Wenne, Farmer WM. 1980. Sea-slug Gastropods. W.M. R. (2006) Introgression and mitochondrial Farmer Enterprises, Tempe, AZ. DNA heteroplasmy in the Baltic populations Fraïsse, C., Belkhir, K., Welch, J. & Bierne, N. of mussels Mytilus trossulus and M. edulis. Mar. (2016) Local interspecies introgression is the Biol., 149, 1371–1385. main cause of extreme levels of intraspeciº»c Knipowitsch, N. (1902) Zoologische ergebnisse di¼ferentiation in mussels. Mol. Ecol. Evol., 25, der russischen expeditionen nach Spitzbergen, 269–286. Mollusca und Brachiopoda. Ann. Mus. Zool. Friele, H. (1879) Catalog der auf der norwegischen Acad. Imp. Sci. St. Pétersbourg, 7, 355–459. Nordmeerexpedition bei Sptizbergen Korshunova, T., Fletcher, K., Picton, B., Lundin, gefundenen Mollusken. Jahrb. Deutsch. K., Kashio, S., Sanamyan, N., Sanamyan, K., Malakozool. Gesellsch., 6, 264–286. Padula, V., Schrodl, M. & Martynov A. (2020a) Gmelin, J. (1791) In: Linnaeus (Ed.) Systema The Emperor Cadlina, hidden diversity and gill Naturae, per regna tria naturae. 13, 1, 6, 3103– cavity evolution: new insights for the taxonomy 3107, 3147–3148. and phylogeny of dorid nudibranchs (Mollusca: González, A., Thompson, P. & Loreau, M. (2018) Gastropoda). Zool. J. Linn. Soc., 189, 762–827. Spatial ecological networks: planning for Korshunova, T., Malmberg, K., Prkić, J., Petani, A., sustainability in the long-term. Curr. Opin. Fletcher, K., Lundin, K. & Martynov, A. (2020b) Environ. Sust., 29, 187–197. Fine-scale species delimitation: speciation in Harrison, R. & Larson, E. (2014) Hybridization, process and periodic patterns in nudibranch introgression, and the nature of species diversity. ZooKeys, 917, 15–50. boundaries. J. Hered., 105, 795–809. Korshunova, T., Martynov, A. & Picton, B. (2017c) Hoeksema, B.W. (2007) Delineation of the Ontogeny as an important part of integrative Indo-Malayan Centre of Maximum Marine taxonomy in tergipedid aeolidaceans Biodiversity: The Coral Triangle. In: Renema (Gastropoda: Nudibranchia) with a description W. (ed.) Biogeography, Time and Place: of a new genus and species from the Barents Distributions, Barriers and Islands, pp 117–178. Sea. Zootaxa, 4324, 1–22. Springer, Dordrecht. Korshunova, T., Martynov, A., Bakken, T. & Picton, ICZN. (1999) International Code of Zoological B. (2017b) External diversity is restrained Nomenclature. The International Trust for by internal conservatism: New nudibranch Zoological Nomenclature, London. mollusc contributes to the cryptic species Jančúchová-Lásková, J., Landová, E. & Frynta, D. (2015) problem. Zool. Scr., 46, 683–692. Experimental crossing of two distinct species of Korshunova, T., Martynov, A., Bakken, T., Evertsen, leopard geckos, Eublepharis angramainyu and J., Fletcher, K., Mudianta, I., Saito, H., Lundin, K., E. macularius: Viability, fertility and phenotypic Schrödl, M. & Picton, B. (2017a) Polyphyly of the variation of the hybrids. PLoS ONE, 10, e0143630. traditional family Flabellinidae a¼fects a major Johnson, R.F. (2010) Breaking family ties: group of Nudibranchia: aeolidacean taxonomic taxon sampling and molecular phylogeny reassessment with descriptions of several of chromodorid nudibranchs (Mollusca, new families, genera, and species (Mollusca, Gastropoda). Zool. Scr., 40, 137–157. Gastropoda). ZooKeys, 717, 1–139.

 $        # $# !"%   %   ./-  |   .

Korshunova, T., Nakano, R., Fletcher, K., Sanamyan, museum collections and modern systematics: N. & Martynov, A. (2019b) First molecular a relict population of the Arctic nudibranch conº»rmation of the presence of Dendronotus Dendronotus velifer G.O. Sars, 1878 in a Swedish primorjensis Martynov, Sanamyan & Korshunova, ºìord. Contrib. Zool., 86, 303–318. 2015 in Japan and new distributional records MacFarland, F.M. (1966) Studies of of Dendronotus species in the North Pacic opisthobranchiate mollusks of the Paciº»c coast (Nudibranchia: Dendronotidae). Venus, 77, 1–14. of North America. Mem. Calif. Acad. Sci., 6, 1–546. Korshunova, T., Picton, B., Furfaro, G., Mariottini, Mallet, J. (2005) Hybridization as an invasion of P., Pontes, M., Prkić, J., Fletcher, K., Malmberg, the genome. Trends Ecol. Evol., 20, 229–237. K., Lundin, K. & Martynov, A. (2019a) Multilevel Mallet, J., Meyer, A., Nosil, P., & Feder, J. (2009) º»ne-scale diversity challenges the ‘cryptic Space, sympatry and speciation. J. Evol. Biol., 22, species’ concept. Sci. Rep., 9, 6732. 2332–2341. Korshunova, T., Sanamyan, N. & Martynov, Marcus, Ev. & Marcus, Er. (1967) American A. (2016b) Morphological and molecular opisthobranch mollusks Part I, Tropical American evidence indicate Dendronotus primorjensis is opisthobranchs, Part II, Opisthobranchs from the a valid species that has priority over D. dudkai Gulf of California. Stud. Trop. Oceanogr., Miami, (Nudibranchia). ZooKeys, 634, 15–28. 6, 1–256. Korshunova, T., Sanamyan, N., Zimina, O., Fletcher, Martinsson, S. & Erséus, C. (2018) Cryptic K. & Martynov, A. (2016a) Two new species and diversity in supposedly species-poor genera a remarkable record of the genus Dendronotus of Enchytraeidae (Annelida: Clitellata). Zool. J. from the North Paciº»c and Arctic Linn. Soc., 183, 749–762. (Nudibranchia). ZooKeys, 630, 19–42. Martynov, A., Fujiwara, Y., Tsuchida, S., Nakano, Kröyer, H.N. (1847) Zoologisches. I. Aus R., Sanamyan, N., Sanamyan, K., Fletcher, K. & Königlichen Musäen. Amtlicher Bericht über Korshunova, T. (2020a) Three new species of die Versammlung Deutscher Naturforscher und the genus Dendronotus from Japan and Russia Ärzte in Kiel, 24, 109–115. (Mollusca, Nudibranchia). Zootaxa, 4747, 495–513. Kumar, S., Stecher, G., &Tamura, K. (2016) MEGA7: Martynov, A., Ishida, Y., Irimura, S., Tajiri, R., Molecular evolutionary genetics analysis O’Hara, T. & Fujita, T. (2015b) When ontogeny version 7.0 for bigger datasets. Mol. Biol. Evol., matters: a new Japanese species of brittle star 33, 1870–1874. illustrates the importance of considering both Lafont, A. (1871) Note pour servir a la faune de la adult and juvenile characters in taxonomic Gironde contenant la liste des animaux marins practice. PLOS ONE, 10, e0139463. dont la présence a été constatée a Arcachon Martynov, A., Lundin, K., Picton, B., Fletcher, pendant les années 1869–70. Act. Soc. Linn. K., Malmberg, K. & Korshunova, T. (2020b) Bordeaux, 28, 237–280. Multiple paedomorphic lineages of soft- Lamb, A. & Hanby, B.P. (2005) Marine Life of the substrate burrowing invertebrates: parallels Paci˜™c Northwest, a Photographic Encyclopedia in the origin of Xenocratena and Xenoturbella. of Invertebrates, Seaweeds and Selected Fishes, PLoS ONE, 15, e0227173. Harbour Publishing, Madeira Park, BC. Martynov, A., Sanamyan, N. & Korshunova, Lin, G., Zhang, F. & Ma, S. (1986) Opisthobranchia. T. (2015a) New data on the opisthobranch In: Chinese animal atlas: Mollusca, volume 3. molluscs (Gastropoda: Opisthobranchia) of Science Publishing Company, China. waters of Commander Islands and Far-Eastern Lundin, K., Korshunova, T., Malmberg, K. & seas of Russia. Conservation of biodiversity of Martynov, A. (2017) Intersection of historical Kamchatka and coastal waters – Proceedings

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

of the XV International Scientiº»c Conference Nakano, R. (2018) Field Guide to Sea Slugs and Petropavlovsk-Kamchatsky, pp. 55–69. Kamchat Nudibranchs of Japan. Bun-ichi Co., Ltd., Press, Petropavlovsk-Kamchatsky. Tokyo. Martynov, A., Sanamyan. N. & Korshunova, NMNH (National Museum of Natural History, T. (2015c) Review of the opisthobranch Smithsonian Institution) (2020) Dendronotus mollusc fauna of Russian Far Eastern seas: elegans Verrill, 1880 Holotype USNM Pleurobranchomorpha, Doridida and 842116. Accesssed through http://n2t.net/ Nudibranchia. Bull. Kamchatka State Technol. ark:/65665/39fc3ac7d-5caa-47ae-b0cb- Univ., 34, 62–87. 60371b864c4d Martynov, A.V. & Korshunova, T.A. (2011) Nordsieck, F. (1972) Die europäischen Opisthobranch Molluscs of the Seas of Russia. Meeresschnecken (Opisthobranchia mit A Colour Guide to their Taxonomy and Biology. Pyramidellidae; Rissoacea), Vom Eismeer bis Fiton, Moscow. Kapverden, Mittelmeer und Schwarzes Meer. Mayer, F. & von Helversen, O. (2001) Cryptic Gustav Fischer Verlag, Stuttgart. diversity in European bats. Proc. R. Soc. B, 268, Nylander, J., Ronquist, F., Huelsenbeck, J. & Nieves- 1825–1832. Aldrey, J. (2004) Bayesian phylogenetic analysis Mayr, E. (1942) Systematics and the Origin of Species of combined data. Syst. Biol., 53, 47–67. Columbia University Press, New York. O’Donoghue, C. (1921) Nudibranchiate Mollusca Mayr, E. (1969) Principles of Systematic Zoology. from the Vancouver Island region. Trans. R. Can. McGraw-Hill, New York. Inst., 13, 147–209. Mayr, E. (1982) The Growth of Biological Thought: O’Donoghue, C. (1922) Notes on the nudibranchiate Diversity, Evolution, and Inheritance. Harvard Mollusca from the Vancouver Island region. III. University Press, Harvard, MA. Records of species and distribution. Trans. R. Mcdonald, G. (1983) A review of the nudibranchs Can. Inst., 14, 145–167. of the California coast. Malacologia, 24, 114–276. Odhner, N. (1907) Northern and Arctic invertebrates McDonald, G. (2009) Nudibranch Systematic Index. in the collection of the Swedish state museum. Second edition. Long Marine Laboratory, UC Det Kongelige Norske Videnskabers Selskabs Santa Cruz. Skrifter, 41, 1–114. McDonald, J., Koehn, R. (1988) The mussels Odhner, N. (1936) Nudibranchia Dendronotacea—A Mytilus galloprovincialis and M. trossulus on revision of the system. Mém. Mus. R. Hist. Nat. the Paciº»c coast of North America. Mar. Biol., Belg. (Ser. 2), 3, 1057–1128. 99, 111–118. Odhner, N. (1939) Opisthobranchiate Mollusca Møller, H.P.C. (1842) Index Molluscorum from the western and northern coasts of Groenlandiæ.C.A. Reitzelii, Hafniae. Norway. Det Kongelige Norske Videnskabernes Mörch, O. (1875) Prodromus Faunæ Molluscorum Selskabs Skrifter, 1, 1–93. Groenlandiæ,. In: Rupert Jones, T. (Ed.) Manual Platania, L., Vodă, R., Dincă, V., Talavera, G., Vila, R. of the Natural History, Geology, and Physics of & Dapporto, L. (2020) Integrative analyses on Greenland and the Neighbouring Regions, 2014 Western Palearctic Lasiommata reveal a mosaic Edition, pp. 124–135. Cambride University Press, of nascent butterÐÑy species. J. Zool. Syst. Evol. Cambridge. Res., doi:10.1111/jzs.12356 Müller, O.F. (1776) Zoologiae Danicae. Prodromus Pola, M. & Stout, C. (2008) Description of the seu animalium Daniae et Norvegiae ingenarum º»rst two tropical Indo-Paciº»c species of characteres, nomina, et synonyma imprimis Dendronotus (Gastropoda: Nudibranchia) popularium. Hallageriis, Havniae. with new data of the poorly known species

 $        # $# !"%   %   ./-  |   .

Dendronotus gracilis Baba, 1949. Zootaxa, 1960, Roginskaya, I.S. (1990) Additional data on the food 45–66. and feeding of Dendronotus robustus Verrill, Pola, M., & Gosliner, T. M. (2010) The º»rst molecular 1879 (Nudibranchia: Dendronotidae) from phylogeny of cladobranchian opisthobranchs Dvinsky Bay of the White Sea. In: Kuznetsov (Mollusca, Gastropoda, Nudibranchia). Mol. A.P. (Eds) Feeding and Bioenergetics of Marine Phyl. Evol., 56, 931–941. Bottom Invertebrates, pp. 93–110, 158. P.P. Pola, M., Rudman, W. & Gosliner, T. (2009) Systematics Shirshov Institute of Oceanology, Academy of and preliminary phylogeny of Bornellidae Sciences of the U.S.S.R., Moscow. (Mollusca: Nudibranchia: Dendronotina) based Roginskaya, I.S. (1997) Predation by a nudibranch on morphological characters with description of Dendronotus robustus from the Sea of Japan on four new species. Zootaxa, 1975, 1–57. Oweniid polychaetes. Opisthobranch Newsl., Potkamp, G. & Fransen, C.H.J.M. (2019) Speciation 23, 21–23. with gene ÐÑow in marine systems. Contrib. Zool., Ronquist, F., Teslenko, M., van der Mark, P., Ayres, 88, 133–172. D.L., Darling, A., Höhna, S., Larget, B., Liu, Pruvot-Fol, A. (1954) Mollusques Opisthobranches. L., Suchard, M.A. & Huelsenbeck, J.P. (2012) Faune de France, 58, 1–460. MrBayes 3.2: E¼º»cient Bayesian phylogenetic Puillandre, N., Lambert, A., Brouillet, S. & Achaz, inference and model choice across a large G. (2012) ABGD, Automatic Barcode Gap model space. Syst. Biol., 61, 539–542. Discovery for primary species delimitation. Rudman, W. (2007) Comment on Dendronotus Mol. Ecol., 21, 1864–1877. robustus from Spitzbergen, Nth Atlantic by Ratnasingham, S. & Hebert, P. (2013) A DNA- Erling Svensen. Message in Sea Slug Forum. Based Registry for All Animal Species: Australian Museum, Sydney. Available from The Barcode Index Number (BIN) System. http://www.seaslugforum.net/º»nd/19087 PLoS ONE, 8, e66213. doi:10.1371/journal. Rybakova, E., Galkin, S., Gebruk, A., Sanamyan, pone.0066213 N. & Martynov, A. (2020) Vertical distribution Reydon, T. & Kunz, W. (2019) Species as natural of megafauna on the Bering Sea slope based entities, instrumental units and ranked on ROV survey., PeerJ, 8, e8628. doi 10.7717/ taxa: new perspectives on the grouping and peerj.8628 ranking problems. Biol. J. Linn. Soc., 126, Sars, G. (1878) Bidrag til kundskaben om Norges 623–636. Arktiske fauna. I. Mollusca regionis Arcticae Robilliard, G. (1970) The systematics and some Norvegiae, oversigt over de i Norges Arktiske region aspects of the ecology of the genus Dendronotus. forekommende. Bløddyr. Universitetsprogram, Veliger, 12, 433–479. Christiania. Robilliard, G. (1972) A new species of Dendronotus Shapiro, B.J., Leducq, J.-B. & Mallet, J. (2016) What from the northeastern Paciº»c with notes on is speciation?. PLoS Genet., 12, e1005860. Dendronotus nanus and Dendronotus robustus Shaw, G. (1991) Chemotaxis and lunge-feeding (Mollusca: Opisthobranchia). Can. J. Zool., 50, behaviour of Dendronotus iris (Mollusca, 421–432. Opisthobranchia). Can. J. Zool., 69, 2805–2810. Robilliard, G. (1975) The nudibranch Dendronotus Stamatakis, A., Hoover, P. & Rougemont, J. (2008) frondosus – one species or four? Festivus, 6, Rapid bootstrap algorithm for the RAxML web- 44–47. servers. Syst. Biol., 75, 758–771. Roginskaya, I.S. (1987) Order Nudibranchia Stanton, D., Frandsen, P., Waples, R., Heller, R., Blainville, 1814. In: Molluscs of the White Sea. Russo, I., Orozco-terWengel, P., Pedersen, C., Keys to the Fauna of USSR, 151, 155–201. Siegismund, H. & Bruford, M. (2019) More grist

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

for the mill? Species delimitation in the genomic Gastropoda: Heterobranchia) with the era and its implications for conservation. description of a new species. J. Mar. Biol. Ass. Conserv. Genet., 20, 101–113. U.K., 97, 303–319. Stout, C., Pola, M. & Valdés, Á. (2010) Phylogenetic Verril, A. (1880) Notice of recent additions to the analysis of Dendronotus nudibranchs with marine Invertebrata, of the northeastern coast emphasis on northeastern Paciº»c species. J. of America, with descriptions of new genera Mollus. Stud., 76, 367–375. and species and critical remarks on others. Stout, C., Wilson, N. & Valdés, Á. (2011) A new Part II.—Mollusca, with notes on Annelida, species of deep-sea Dendronotus Alder & Echinodermata, etc., collected by the United Hancock (Mollusca: Nudibranchia) from States Fish Commission. Proc. US Nat. Hist. California, with an expanded phylogeny of the Mus., 3, 356–405. genus. Invertebr. Syst., 25, 60–69. Verrill, A. (1870) Contributions to zoology from the Swennen, C. (1961) Data on distribution, Museum of Yale College. No.8.—Descriptions reproduction and ecology of the nudibranchiate of some New England Nudibranchiata. Am. J. molluscs occurring in the Netherlands. Neth. J. Sci. Arts (Ser. 2), 50, 405–408. Sea Res., 1, 191–240. Vervoort, W. (1942) Northern Hydroida in the Talavera, G. & Castresana, J. (2007) Improvement Collections of the Rijksmuseum van Natuurlijke of phylogenies after removing divergent and Historie and the Zoological Museum at ambiguously aligned blocks from protein Amsterdam, with notes on their distribution. sequence alignments. Syst. Biol., 56, 564–577. Zool. Meded,, 23, 275–312. Thollesson, M. (1998) Discrimination of two Wägele, H., Willan, R. (2000) Phylogeny of the Dendronotus species by allozyme electro- Nudibranchia. Zool. J. Linn. Soc., 130, 83–181. phoresis and the reinstatement of Dendronotus Wiemers, M. & Fiedler, K. (2007) Does the DNA lacteus (Thompson, 1840) (Nudibranchia, barcoding gap exist? – a case study in blue Dendronotoidea). Zool. Scr., 27, 189–195. butterÐÑies. Front. Zool., 4, 8. Thompson, T. & Brown, G. (1984) Biology of Willis, S. (2017) One species or four? Yes!…and, no. Opisthobranch Molluscs (vol. 2). The Ray Society Or, arbitrary assignment of lineage to species Publishing, London. obscures the diversiº»cation processes of Thompson, W. (1840) Contributions towards a Neotropical º»shes. PLoS ONE, 12, e0172349. knowledge of the Mollusca Nudibranchia and Wobber, D. (1970) A report on the feeding of Mollusca Tunicata of Ireland, with descriptions Dendronotus iris on the anthozoan Cerianthus sp. of some apparently new species of invertebrata. from Monterey Bay, California. Veliger, 12, 383–387. Ann. Mag. Nat. Hist., 5, 84–102, Wollscheid-Lengeling, V., Boore J., Brown, Valdés, Á. & Bouchet, P. (1998) Naked in toxic ÐÑuids: W. & Wägele, H. (2001) The phylogeny of a nudibranch mollusc from hydrothermal vents. Nudibranchia (Opisthobranchia, Gastropoda, Deep-sea Res. Pt II, 45, 319–327. Mollusca) reconstructed by three molecular Valdés, Á., Lundsten, L. & Wilson, N. (2018) markers. . Org. Divers. Evol., 1, 241–256. Five new deep-sea species of nudibranchs Yale Peabody Museum. (2020) Dendronotus (Gastropoda: Heterobranchia: ) elegans Verrill, 1880 YPM IZ 010761.GP. Accessed from the Northeast Paciº»c. Zootaxa, 4526, 4, through https://collections.peabody.yale.edu/ 401–433. search/Record/YPM-IZ-010761.GP Valdés, Á., Murillo, F., McCarthy, J. & Yedinak, N. Yang, Z. (2015) The BPP program for species tree (2017) New deep-water records and species estimation and species delimitation. Curr. Zool., of North Atlantic nudibranchs (Mollusca, 61, 854–865.

 $        # $# !"%   %   ./-  |   .

Yoder, A., Olson, L., Hanley, C., Heckman, K., Zachos, F.E. (2018) Mammals and meaningful Rasoloarison, R., Russell, A., Ranivo, J., taxonomic units: the debate about species Soarimalala, V., Praveen Karanth, K., concepts and conservation. Mammal Rev., 48, Raselimanana, A. & Goodman, S. (2005) A 153–159. multidimensional approach for detecting species Zachos, F.E., Apollonio, M., Bärmann, E.V., patterns in Malagasy vertebrates. In: Hey, J., Fitch, Festa-Bianchet, M., Göhlich, U., Habel, J.C. W.M., Ayala, F.J., (Eds) Systematics and the origin et al. (2013) Species inÐÑation and taxonomic of species: on Ernst Mayr’s 100th anniversary, artefacts—A critical comment on recent pp. 203–228 The National Academies Press, trends in mammalian classiº»cation. Mamm. Washington, D.C.. Biol., 78, 1–6. Zachos, F.E. (2016) Species concepts in biology. Historical Development, Theoretical Foundations Û  : 02 MÛ 2020 |    and Practical Relevance. Springer Nature ÛÛÚ : 21 JÁ 2020 Switzerland, Cham.  : .í. Ø

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   H3 KP871661 MN138207 MH756144 – – 28S – – 16S MN138057– – Û KP871637KX788135KX058081 KP871685 KX788123 KX058121 – MN138293 KX788114– KX058093 – KX058109 KM396976KY391832 KM397056 KY391852MT654640 KM397017KM396996 MT655312 KY391870 KM397111 KM397078 MT655541 – KM397037 KM397087 KM396997 KM397079 KM397038 – Locality USA, North Atlantic? North USA, AF249800Canada: Newfoundland KU695598 AF249252 – AY427450 – Voucher CASIZ192342ZMMU:Op-566LACM:2004-2.2ZMMU:Op-561 – Washington USA: ZMMU:Op-562 California USA: MIMB36291.2 Sea Laptev Russia: 3554LACM Sea Laptev Russia: – KX788140 Mexico KX788141 KX788129 KX788130 KX788118 KX788119 – – ZMMU:Op-330ZMMU:Op-331– Kamchatka Russia: ZMMU:Op-380 Kamchatka Russia: ZMMU:Op-588NF09 Norway KM396999 Norway ZMMU:Op-713 KM397000ZMMU:Op-269 KM397081 KM397082 Greenland ZMMU:Op-279 sea White Russia: KM397040 KM397041ZMMU:Op-274.1 Sea Barents Russia: KM397102 ZMMU:Op-274.2 Sea Barents Norway: KM397103 ZMMU:Op-274.3 Sea Barents Norway: KM396995 KC660037ZMMU:Op-270 Sea Barents Norway: KM397077 KM396993 KC611293 sea White Russia: KM397036 KM396994 KM397076 KC660021 KM397091 KM397075 KM397034 – KM397035 KM397089 – Appendix A1  List of specimens used for the molecular analyses Species name Crosslandia viridis Crosslandia albus Dendronotus albus Dendronotus arcticus Dendronotus arcticus Dendronotus arcticus Dendronotus claguei Dendronotus Dendronotus dalli Dendronotus dalli Dendronotus dalli Dendronotus frondosus Dendronotus frondosus Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus (as elegans Dendronotus ”) “ niveus Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus (as elegans Dendronotus ”) “ niveus Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus

 $        # $# !"%   %   ./-  |   . H3 – – – – 28S MT655540 16S – Û KC660036KM396998 KC611292MT654635 KM397080MT654638 KC660020MT654639 MT655308 KM397039 MT655311 – MT655536 – MT655539 MN138334MN138331 MN138100MN138332 MN138097 MN138174MN138333 MN138098 MN138171 – MN138330 MN138099 MN138172 – KY391823 MN138096 MN138173KY391826 – KY391824 KY391842 MN138170KY391821 – KY391845KY391837 KY391843 – KY391860KX058083 KY391840 KY391863KX058082 KY391839 KY391861MN808558 HM162631 – KY391858 GU339188 – MN808559 KY391857 MN811023 – KX058096 – KX058095 MN811024 MN811017 – HM162537 KX058110 – – Locality Russia: White sea White Russia: sea White Russia: Russia: White sea White Russia: Sea Kara Russia: Sea Kara Russia: Sea Kara Russia: Sea Kara Russia: Japan Japan Voucher WS1102 WS2005 ZMMU:Op-715ZMMU:Op-716ZMMU:Op-717 sea Laptev Russia: sea Laptev Russia: sea Laptev Russia: WS9130 WS9123 WS9124 WS9125 IE211 ZMMU:Op-554ZMMU:Op-581ZMMU:Op-579 Norway ZMMU:Op-578 UK ZMMU:Op-555 UK CASIZ:174471 Norway LACM:174194 Norway JAMSTEC No. 1160047463 Washington USA: JAMSTEC Washington USA: No. 1160047475 ZMMU:Op-284.1ZMMU:Op-284.2 Kamchatka Russia: Kamchatka Russia: KC660026 KC660025 KC611285 KC611286 KC660014 KC660013 – – Species name elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus elegans Dendronotus elegans Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus ”) “ niveus (as Dendronotus elegans Dendronotus “) “ niveus (as Dendronotus elegans Dendronotus “) “ niveus (as Dendronotus ” “ elegans Dendronotus “) “ niveus (as Dendronotus europaeus Dendronotus europaeus Dendronotus europaeus Dendronotus europaeus Dendronotus europaeus Dendronotus iris Dendronotus iris Dendronotus jamsteci Dendronotus jamsteci Dendronotus kalikal Dendronotus kalikal Dendronotus

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   H3 MN138229 MN138230 – – HM162535 JN869430 28S 16S Û KX788144KY391830KY391838 KX788111KY391829 KY391849KY391831 KY391851KM396973 KX788121 KY391848 KY391867MT654636 KY391850 KY391869 KM397057 – HQ225828 KY391866 MT655309 – KY391868 KM397014 – HQ225829 MT655537 – KT031812 KM397105 – HM162708 – JN869451 KT031825 HM162629MK302459 JN869407 KT031842 – MK302454 – MN138263 MK302466 – Locality Russia: Kuril Islands Kuril Russia: Islands Kuril Russia: MN138314 MN138315 MN138079 MN138080 MN138153 MN138154 – KamchatkaRussia: KamchatkaRussia: KX058077 KX058078 KX058117 KX058118 KX058089 KX058090 KX058105 KX058106 Voucher ZMMU:Op-284.3ZMMU:Op-283 Kamchatka Russia: ZMMU:Op-285ZMMU:Op-349.1 Kamchatka Russia: ZMMU:Op-349.2 Bering Strait Russia: ZMMU:Op-657 Kamchatka Russia: KM396988 Kamchatka Russia: WS9131 WS9132 KC660024 Islands Kuril Russia: KM397070 KC660027ZMMU:Op-245 KM396986ZMMU:Op-565 KC611284 KM396987 KM397029 KC611287ZMMU:Op-584 Kamchatka Russia: MK302458 KM397068ZMMU:Op-587Washington USA: KM397069 KC660012 – ZMMU:Op-586 Norway KC660015 MK302453 KM397027ZMMU:Op-585 Russia KM397028ZMMU:Op-335 KC660032 KC660044 Norway KC660047 – MK302465ZMMU:Op-665 Norway KM397109 KC611288 Russia SIO-BIC M12133 – sea Laptev Russia: ZMMU:Op-419ZMMU:Op-420 KC660016 California USA: W196 Sea of Russia: JapanCASIZ:179492 KC660048 Sea of Russia: JapanCASIZ:179493ZMMU:Op-568 Philippines KX672010ZMMU:Op-659 KX672011 Philippines IE251 Kamchatka Russia: IE252 KX672008Washington USA: ZMMU:Op-343 KX672009 KX672006 KX788138 Sea Barents Russia: KX672007 – KX788126 – KM397002 KX788116 KM397084 – KM397043 KM397106 T A1 T List of) specimens used for the molecular analyses ( cont. Species name Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kalikal Dendronotus kamchaticus Dendronotus kamchaticus Dendronotus lacteus Dendronotus lacteus Dendronotus lacteus Dendronotus lacteus Dendronotus lacteus Dendronotus sp. nordenskioeldi Dendronotus nov. patricki Dendronotus primorjensis Dendronotus primorjensis Dendronotus primorjensis Dendronotus Cabangus regius Cabangus regius robilliardi Dendronotus robilliardi Dendronotus robilliardi Dendronotus robilliardi Dendronotus robustus Dendronotus

 $        # $# !"%   %   ./-  |   . H3 MN138274 MN138279 MN138290 – – – – – MN138232 MN138231 MN138233 MN138226 MN138227 MN138227 MN138228 KJ509924 HM162554 HM162553 28S MN138203 – – 16S MN138078 MN138152 Û KX058084KX058085MN808564 GU339191MN138347 KX058122MF685027 MN811029 KX058097 MN138113 KX058098MN138352 KY996407 MN811020 HQ267091 MK302460 MN138187 KX058111 MN138118 MK302461 – MT654641 MK302455MT654642 – MN138192 MT654643 MT655313 MK302467MT654644 MT655314MT654645 MT655315 – MT655542 MN138317 MT655316 MT655543 MN138316 MT655317 MT655544 MN138318 MN138082MN138311 MN138081MN138312 MN138083 MN138156 MN138313 MN138076 MN138155 MN808562 MN138077 MN138157 MN138150 MN811027 MN138151 KJ486734HM162722 MN811019 HM162721 KJ486756 HM162646 HM162645 – – – Locality USA: Oregon USA: – Sea Kara Russia: Sea Kara Russia: Sea Kara Russia: Sea White Russia: Sea White Russia: Sea White Russia: Voucher ZMMU:Op-344ZMMU:Op-390.5LACM:174861 Sea Barents Russia: Sea Barents Russia: LACM:174863ZMMU:Op-699 Alaska USA: WS9098 Alaska USA: KM397003ZMMU:Op-348Washington USA: KM396968ZMMU:Op-546IE182 KM397085 Sea Kara Russia: KM397051ZMMU:Op-660 Sea Laptev Russia: CPIC-01179 KM397044 KM397009Washington USA: – KM397119 KY996409 Canada: British Columbia MN138363 KY996405 MN138129 MK302462 – KSNHM: OP0485ZMMU:Op-700 Japan ZMMU:Op-571 Mn33142BELUM Kamchatka Russia: CASIZ:177578 Sea Barents Russia: UK CASIZ:176812 Philippines MN808561 Portugal KX788145 MN811026 KX788133 MN811018 KX788113 – – Species name Dendronotus robustus Dendronotus robustus Dendronotus rufus Dendronotus rufus Dendronotus subramosus Dendronotus subramosus Dendronotus velifer Dendronotus velifer Dendronotus velifer Dendronotus venustus Dendronotus venustus Dendronotus sp. nov. yrjargul Dendronotus sp. nov. yrjargul Dendronotus ZMMU: Op-718 sp. nov. yrjargul Dendronotus NTNU-VM-76302 sp. nov. yrjargul Dendronotus NTNU-VM-76306 sp. nov. yrjargul Dendronotus Sea Norwegian Norway: Sea Norwegian Norway: NTNU-VM-76308 sp. nov. yrjargul Dendronotus Sea Norwegian Norway: NTNU-VM-76305 sp. nov. yrjargul Dendronotus Sea Norwegian Norway: WS9116 sp. nov. yrjargul Dendronotus Sea Norwegian Norway: WS9113 sp. nov. yrjargul Dendronotus WS9117 sp. nov. yrjargul Dendronotus WS9103 sp. nov. yrjargul Dendronotus WS9126 zakuro Dendronotus WS9127 zakuro Dendronotus coronata Doto tuberculata Doto arborescens Marionia blainvillea Marionia

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-   H3 JN869439 JN869437 – HM162534 JN869432 JN869443 – EF133468 HM162548 28S – 16S HM162628– – Û JN869456JN869454MT654637 JN869413– JN869411– MT655310 – – MT655538 JN869459KX889746AY345035 JN869416HM162716 MK100991KX788134 AY345035 – HM162641 – KX788122 – – KX788132 – Locality – Voucher CASIZ:176362CASIZ:177540ZMMU: Op-664 South Africa CASIZ:174989 Philippines CASIZ:174988 Sea of Russia: Japan China China CASIZ184317SRR3726701– CASIZ:176219 Philippines ZMMU:Op-572 – South Africa Norway T A1 T List of) specimens used for the molecular analyses ( cont. Species name Notobryon thompsoni Notobryon wardi Notobryon orientalis Pseudobornella orientalis Pseudobornella orientalis Pseudobornella Scyllaea pelagica Scyllaea fulva Scyllaea ˜™mbria Tethys nilsodhneri Tritonia plebeia Tritonia

 $        # $# !"%   %   ./-  |   .

C. regius

P. orientalis

D. velifer 13,9 1 D. patricki 3,5 17,9 18,4

D. robustus 16,2 16,0 16,0 18,0 17,9

D. robilliardi 12,1 13,71.41 12,2 13,1 15,7 16,8

D. iris 12,5 13,90.63 14,8 14,0 14,7 18,0 18,3

D. jamsteci 12,0 11,20.46 12,6 15,0 14,6 14,3 17,9 16,9 6,5 9,6 10,4 13,5 14,4 D. primorjensis 0.10

D. frondosus 8,4 8,70.46 10,5 9,9 13,6 14,7 14,5 13,7 16,4 17,1

D. venustus 9,5 10,80.19 11,0 10,7 9,4 11,9 13,8 13,1 12,6 17,4 16,5

D. yrjargul sp. nov. 2,10.39 9,1 10,6 10,6 11,3 9,7 11,9 14,6 14,0 13,6 18,0 16,2

D. kalikal 0.80

D. nordenskioeldi sp. nov. 7,9 11,1 10,6 11,0 12,6 14,1 10,2 12,5 12,8 14,5 15,4 15,3 16,4 17,2

D. europaeus 7,3 8,30.20 11,4 11,1 12,0 12,1 13,3 11,7 12,5 12,7 14,3 16,5 16,0 16,8 17,5

D. rufus 4,6 8,40 8,3 12,2 11,9 12,3 12,6 14,2 12,7 13,8 13,7 14,6 16,8 16,8 18,3 16,7

D. lacteus 10,1 8,80.61 8,3 9,0 9,7 9,9 10,2 10,9 12,2 11,3 12,4 11,5 13,5 14,7 14,2 15,9 16,4

D. elegans 7,6 9,40.59 8,5 9,5 8,9 12,0 11,9 12,2 13,3 13,7 12,5 13,6 13,3 15,1 15,9 16,0 16,9 16,7

D. dalli 8,7 8,61.45 8,7 8,4 8,2 9,1 9,6 10,0 10,0 10,0 11,0 10,3 11,9 12,4 14,5 14,7 14,0 16,5 16,6

D. zakuro 7,0 8,51.37 8,9 9,4 8,6 9,2 8,8 11,4 12,1 10,2 11,6 11,6 11,9 12,8 12,4 15,7 15,5 14,8 16,1 18,1

D. kamchaticus 9,3 7,81.15 9,6 9,7 10,5 10,0 8,9 9,9 10,0 9,7 10,6 10,7 12,1 10,6 11,6 12,9 13,2 14,4 13,1 16,5 17,7

D. arcticus 12,0 11,12.34 10,2 13,9 11,2 12,3 11,5 10,9 12,3 11,0 11,7 11,5 9,7 11,4 9,5 12,9 13,8 16,2 16,0 15,6 17,3 16,8

D. subramosus 9,4 11,50.67 11,1 9,7 13,1 9,9 13,2 12,6 10,7 12,7 10,3 11,1 10,4 11,8 12,1 10,7 13,7 13,5 15,3 16,4 14,9 16,7 17,9 1.02 9,4 11,5 12,0 11,1 11,19,7 9,3 10,213,1 7,8 13,99,9 7,0 9,6 11,213,2 8,5 9,7 12,3 8,7 12,6 10,5 8,9 11,5 9,410,7 8,6 10,0 10,9 8,7 8,6 7,6 12,7 8,9 9,4 12,3 8,4 9,2 10,1 9,9 8,5 8,210,3 8,8 8,8 11,0 9,511,1 9,1 10,0 4,6 8,3 11,7 11,4 8,910,4 8,4 9,6 11,5 9,7 9,011,8 7,3 12,0 10,6 12,1 8,3 9,7 9,7 10,2 10,0 8,3 10,0 10,7 11,9 12,2 12,2 11,6 7,9 9,9 11,4 10,2 10,0 11,1 – 12,3 11,9 13,3 12,0 11,1 11,7 10,9 11,0 11,7 12,6 10,6 12,4 12,1 11,5 11,5 9,1 12,6 2,1 12,4 12,3 12,3 9,5 10,6 14,1 10,8 10,7 8,4 13,5 12,5 14,2 15,0 14,2 17,0 18,2 D. albus 12,1 11,410,7 12,1 9,513,7 11,6 10,6 12,9 11,0 11,913,5 11,6 13,7 13,8 10,3 12,8 12,2 12,5 12,9 11,9 11,3 14,2 12,4 13,6 13,3 12,4 12,7 12,4 13,3 14,1 11,7 13,8 11,5 14,1 10,2 12,5 13,7 10,6 10,7 12,5 12,7 11,3 11,0 13,5 12,8 9,7 8,7 10,7 12,5 10,5 6,5 9,4 11,9 9,6 11,9 9,9 12,0 13,6 10,4 13,5 11,2 12,6 12,5 13,9 12,1  A2  nov in species of and Cabangus gen. Pseudobornella, distances (%) for Û marker genetic Dendronotus, in bold) and intergroup (highlighted Mean intragroup albus D. subramosus D. arcticus D. kamchaticus D. zakuro D. dalli D. elegans D. lacteus D. rufus D. europaeus D. nordenskioeldi D. nov. sp. kalikal D. nov. sp. yrjargul D. venustus D. frondosus D. D. primorjensis D. jamsteci D. iris D. robilliardi D.

 $        # $# !"%   % Ø  À Ø  Á  ÚÛ  ÚØ |  ./-  

C. regius 0

P. orientalis 15,6 17,6

D. velifer 0.21

D. patricki 11,4 10,0 17,4 17,9

D. robustus 0.83

D. robilliardi

D. iris

D. jamsteci

D. primorjensis

D. frondosus

D. venustus

D. yrjargul sp. nov.

D. kalikal

D. nordenskioeldi sp. nov.

D. europaeus

D. rufus

D. lacteus

D. elegans

D. dalli

D. zakuro

D. kamchaticus

D. arcticus

D. subramosus 16,4 16,014,9 14,4 15,6 15,516,7 13,1 14,7 17,3 15,9 14,817,9 16,5 14,7 14,0 16,8 16,1 16,0 17,7 16,8 16,5 14,2 16,5 18,1 16,9 16,8 15,4 16,6 15,9 16,0 15,0 16,7 15,3 18,3 14,0 16,4 16,8 14,2 13,1 16,7 16,4 13,6 17,5 14,5 17,0 12,6 17,2 13,9 18,0 13,7 18,2 17,4 14,6 13,5 16,2 14,0 16,4 14,3 16,5 12,2 17,9 14,7 17,1 16,0 17,9 13,1 11,4 18,4 18,0 16,0 – 15,7 16,9 10,0 18,3 18,0 7,2 17,4 7,2 16,8 16,3 17,9 16,3 15,6 17,9 17,2 – 17,2 17,6 16,9 16,9 D. albus 15,3 16,2 13,2 15,7 14,5 15,1 13,5 14,6 14,3 14,5 14,2 14,6 13,8 14,7 14,4 15,0 14,8 13,7 16,2 ) ( cont. D. patricki D. velifer D. orientalis P. C. regius T A2 T nov , and Cabangus gen. in species of Pseudobornella distances (%) for coi marker genetic Dendronotus, in bold) and intergroup (highlighted Mean intragroup robustus D.

 $        # $# !"%   %