Eubranchidae Odhner, 1934

Family Eubranchidae Odhner, 1934 View in CoL

( Figs 1, 2; Table 4)

Odhner 1934: 282.

Martynov 1998: 763–77.

Korshunova et al. 2017c, family Eubranchidae validity confirmed: 9, 14, 17, 18.

Korshunova et al. 2020b: 21–3, 39.

Diagnosis: Body wide to narrow. Notal edge completely reduced. Cerata non-elevated, numerous-to-few per row. Ceratal rows branched or simple. Rhinophores commonly smooth, rarely annulate or papillate. Anus acleioproctic, very rarely cleioproctic. Masticatory edges of jaws commonly bear single row of distinct denticles, commonly partly sharpened to compound. Radula formula 1.1.1, which is unique among members of the superfamily Fionoidea . Central teeth usually with small cusp, rarely massive, compressed or not compressed by adjacent lateral denticles. Lateral teeth thin, characteristically broadened laterally, commonly smooth, with or without strongly reduced denticles. Distal receptaculum seminis, or it is reduced in several taxa, absence of receptaculum is correlated with presence of a broad lobe-shaped penis, or with a special system of copulative stylets. Vas deferens usually short to moderate, prostate indistinct to distinct. Supplementary gland present; inserts into penis in absolute majority of taxa or, very rarely, in vas deferens. Massive external permanent penial collar absent. Penis internal, commonly armed, rarely unarmed. Three main patterns of copulative architecture: usual not considerably widened, commonly conical penis with a hollow stylet, or a broad lobe with numerous small spines, or a special system of macro- and microstylets in several genera, in which case penis is usually reduced.

Genera included: Aenigmastyletus Martynov, 1998 , reinstated, Amphorina Quatrefages, 1844 , validity confirmed in Korshunova et al. (2020b) and herein, Annulorhina Rao, 1968 , reinstated, Capellinia Trinchese, 1873 , validity confirmed herein, Corruptobranchus Martynov and Korshunova, 2025, Dunga Eliot, 1902 , reinstated, Eubranchopsis Baba, 1949 , reinstated, Eubranchulus Martynov and Korshunova, 2025, Eubranchus Forbes, 1838 , restricted only to type species E. tricolor in Martynov (1998), Korshunova et al. (2020b), and herein, Galvinella Eliot, 1907 , reinstated, Karavellia Martynov and Korshunova, 2025, Leostyletus Martynov, 1998 , validity confirmed herein, Longibranchus Martynov and Korshunova, 2025 , Nihonbranchus Martynov and Korshunova, 2025, Nudibranchus Martynov, 1998 , reinstated, and Produnga Martynov, 1998 , reinstated. See brief genus-level diagnoses below.

Remarks: The family Eubranchidae represents an outstandingly unique taxon within the superfamily Fionoidea , because on one hand it clearly morphologically represents the only survived fionoidean family with the ancestral triserial radula, but on the other hand, Eubranchidae are robustly placed within the otherwise uniserial superfamily Fionoidea as sister to the majority of other families (except Abronicidae ) according to the molecular phylogenetic data ( Figs 1, 2; see Synopsis above). Thus the mutilevel, complex nature of the family Eubranchidae , which from one view is definitely more morphologically ‘related’ to other triserial superfamilies, but from the viewpoint of the particular, otherwise exclusively uniserial, superfamily Fionoidea ( Fig. 2) is the self-evident example that any lumping approach in taxonomy will undermine the correct presentation of the underlying complex evolutionary pathways.

When pan-lumping logic is arbitrarily applied at the family-level, the remarkable genus-level diversity of external morphology, radula, and reproductive apparatus, which has previously been documented in detail ( Martynov 1998, 2005, Korshunova et al. 2020b) within the family Eubranchidae , is fundamentally wiped out under the diagnoses-less and extremely heterogenous so-called genus ‘ Eubranchus ’. On the contrary, for example, every small detail in almost all possible combinations of external, radular, and reproductive system morphology are employed to produce numerous genera within the family Facelinidae , commonly with a few species per genus or monotypic (WoRMS 2024). In the course of our morphological revision of the family Eubranchidae , all major groups have been revealed and clearly recognized taxonomically. For example, ceratal branch morphology, radular patterns, and morphology of the copulative apparatus, especially the latter, as in the family Facelinidae , may be dramatically different between various Eubranchidae genera ( Martynov 1998). In addition, the molecular data ( Korshunova et al. 2020b), which show paraphyly of some even narrowly defined genera, by no means imply that all Eubranchidae diversity must be lumped under the so-called ‘single genus “ Eubranchus ”’. In that respect, it has previously been shown in detail that the type species of the genus Eubranchus , E. tricolor , fundamentally and very distinctly differs from all known representatives of the family in the possession of strongly branched rows of the digestive gland and placement of the supplementary gland in the male reproductive system, which inserts into the vas deferens and not into the penis ( Martynov 1998, Korshunova et al. 2017c, 2020b, Martynov and Korshunova 2025). This combination of characters is unique for the single true genus Eubranchus , and, therefore, none of the more than 50 species currently assigned to the alleged ‘genus Eubranchus ’ can be reliably assigned to the real, proper genus Eubranchus (restricted to only the type species E. tricolor ), the stem-genus name of the family Eubranchidae . These morphological patterns are drastically different from all other Eubranchidae , and cannot be ignored. Therefore, at the present time, a maximally possible, major genus-level Eubranchidae diversity is delineated, but an exhaustive species-level synopsis is not given, because several undescribed genera still need to be separated. However, any of the Eubranchidae taxa that cannot be assigned to any of the currently recognized genera, can also not be considered as belonging to the genus Eubranchus proper and must be designated in WoRMS and other sources only as ‘ Eubranchus ’ in quotation marks.

Therefore, the previous careful fine-scale morphological revision of genus-level taxa within the family Eubranchidae ( Martynov 1998) cannot be disregarded. While for several new or uncertain species the genus-level position still needs to be assigned (including separation of more new genera), it is urgently necessary to return a finely differentiated taxonomy to the family Eubranchidae , because currently the majority of the disparate diversity within Eubranchidae is unfortunately synonymized with the extremely heterogenous genus ‘ Eubranchus ’, where some genera are chaotically retained as valid, others are not, thus profoundly ignoring previous well-documented morphological and molecular differences ( Martynov 1998, 2005, Martynov and Korshunova 2011, Korshunova et al. 2017c, 2020b). Especially remarkable is the diversity of the uniquely constructed, strongly differentiated copulative stylets (in length, proportion, and number) along with a reduction of the true penis within the genera Aenigmastyletus , Dunga , and Leostyletus (see: Martynov 1998, 2005). Such specially differentiated stylets and the absence of a true penis are an absolutely unique pattern among all families of the suborder Aeolidacea , and, therefore, to include these taxa in the super-heterogenous, pan-lumping genus ‘ Eubranchus ’ sensu latissimo , will not merely mask that overwhelming diversity, but intentionally conceal it. Accordingly, the following genera are reinstated or confirmed as valid, because main diagnoses have been already provided in Martynov (1998) and Korshunova et al. (2020b) and, therefore, are presented here in a brief form:

Eubranchus Forbes, 1838 (restricted to only its type species E. tricolor Forbes, 1938 and any potential closely related hidden diversity). Genus diagnosis is ‘ceratal rows strongly branched, not in regular rows, rhinophores smooth, central teeth of radula with compressed cusp, prostate not distinct, supplementary gland inserts into vas deferens, penis conical, unarmed, distal seminal receptaculum present’.

Nudibranchus Martynov, 1998 View in CoL , reinstated [restricted to the type species N. exiguus (Alder and Hancock, 1848) View in CoL , including hidden diversity]. Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores smooth, central teeth of radula with non-compressed cusp, prostate not distinct, penis conical, usual penial stylet present, distinct, supplementary gland inserts into penis, distal seminal receptaculum present’, for other Nudibranchus species in Martynov (1998) previously included in the genus and potential species yet to be described, several more genera need to be separated.

Produnga Martynov, 1998 View in CoL , reinstated [ type species P. rubropunctata (Edmunds, 1969) also including potential hidden diversity]. Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores papillate, central teeth of radula with non-compressed cusp, prostate not distinct, usual penial stylet absent, instead there is a broad, flattened, partly folded penial lobe with numerous small spines with broadened base, supplementary gland inserts into penis, seminal receptaculum absent’.

Annulorhina Rao, 1968 View in CoL , reinstated ( type species A. mandapamensis Rao, 1968 View in CoL ) also includes potential hidden diversity. Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores annulate, central teeth of radula with non-compressed cusp, prostate not distinct, usual penial stylet absent, instead there is a broad penial lobe with numerous small spines, supplementary gland inserts into penis, seminal receptaculum absent’.

Dunga Eliot, 1902 View in CoL , reinstated ( type species D. nodulosa Eliot, 1902 ) also includes potential hidden diversity. Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores smooth, central teeth of radula with very strong non-compressed cusp, prostate not distinct, usual penis and penial stylet absent, instead there is a complex system of up to around 30 macrostylets, supplementary gland inserts into copulative atrium with stylets, seminal receptaculum absent’.

Aenigmastyletus Martynov, 1998 View in CoL , reinstated ( type species A. alexeii Martynov, 1998 View in CoL ) also includes potential hidden diversity. Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores smooth, central teeth of radula with moderate non-compressed cusp, prostate not distinct, usual penis and penial stylet absent, instead there is a complex system of two macrostylets inserted into each other in a partly spiral structure and up to about thirty microstylets, supplementary gland inserts into copulative atrium with stylets, seminal receptaculum absent’.

Leostyletus Martynov, 1998 View in CoL , validity confirmed ( type species L. pseudomisakiensis Martynov, 1998 View in CoL ), besides the type species, also includes Leostyletus misakiensis (Baba, 1960) View in CoL . Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores smooth, central teeth of radula with moderate non-compressed cusp, prostate not distinct, usual penis and penial stylet absent, instead there is a complex system of two largely straight macrostylets separated from each other and up to around six microstylets, supplementary gland inserts into copulative atrium with stylets, seminal receptaculum absent’.

Galvinella Eliot, 1907 , reinstated ( type species G. antarctica Eliot, 1907 ) also includes hidden diversity in Antarctic waters, potentially G. glacialis Thiele, 1912 View in CoL and G. adarensis ( Odhner, 1934) . Genus diagnosis (restricted to Antarctic waters) is ‘up to six pre-anal ceratal rows, rhinophores smooth, central teeth of radula with compressed cusp, prostate moderately distinct, penis conical, usual penial stylet present, but difficult to detect, indistinct, supplementary gland inserts into penis, distal seminal receptaculum present’.

Capellinia Trinchese, 1873 View in CoL , validity confirmed ( type species C. doriae Trinchese, 1874 View in CoL ) besides the type species, also includes at least Capellinia fustifera (Loven, 1846) View in CoL , and Capellinia vittata (Alder and Hancock, 1842) comb. nov. Genus diagnosis is ‘up to commonly two, rarely more pre-anal ceratal rows, rhinophores smooth, central teeth of radula with compressed cusp, prostate distinct, penis conical, usual penial stylet present, distinct, supplementary gland inserts into penis, distal seminal receptaculum present’.

Amphorina Quatrefages, 1844 , validity confirmed and species composition restricted in Korshunova et al. 2020b and herein [ type species A. farrani (Alder and Hancock, 1844) (= A. alberti Quatrefages, 1844 )]. Genus diagnosis is ‘up to six pre-anal ceratal rows (commonly up to four), rhinophores smooth, central teeth of radula with compressed cusp, prostate readily distinct, penis conical, usual penial stylet present, distinct, supplementary gland inserts into penis, distal seminal receptaculum present’, including at least the following species: A. andra Korshunova et al., 2020 , reinstated, A. farrani (Alder and Hancock, 1844) , A. linensis (Garcia-Gomez et al., 1990) , A. pallida (Alder and Hancock, 1842) , and A. viriola Korshunova et al., 2020 , reinstated in its original sense, and? Amphorina amazighi (Tamsouri et al., 2015) comb. nov..

Corruptobranchus Martynov and Korshunova, 2025, [ type species C. odhneri (Derjugin and Gurjanova, 1926) ]. Genus diagnosis is ‘up to 4–6 pre-anal ceratal rows, rhinophores smooth, central teeth of radula with non-compressed to weakly compressed cusp, prostate moderately distinct, penis conical, usual penial stylet present, indistinct, supplementary gland inserts into penis, distal seminal receptaculum present’, besides the type species, includes at least C. odhneri (Derjugin and Gurjanova, 1926) , C. malakhovi (Ekimova et al., 2021) , and C. sanjuanensis (Roller, 1972) .

Eubranchopsis Baba, 1949 , reinstated ( type species Eubranchopsis virginalis Baba, 1949 ). Genus diagnosis is ‘two pre-anal ceratal rows, cerata with conspicuous soft spines and tubercles, rhinophores smooth, central teeth of radula with compressed cusp, prostate distinct, penis conical, usual penial stylet present, distinct, supplementary gland inserts into penis, distal seminal receptaculum present’, hidden diversity exists.

Longibranchus Martynov and Korshunova, 2025 [ type species L. putnami (Fernandez-Simon and Moles in Cunha et al., 2023)]. Genus diagnosis is ‘one cerata per row, in majority of cases, c. 3–5 and more (depending on length) pre-anal rows with a single cerata per row, rhinophores smooth, central teeth of radula with moderately compressed cusp, prostate distinct, penis conical, usual penial stylet present, distinct, partly fold-like, supplementary gland inserts into penis, distal seminal receptaculum present’, apart from the type species, there is potential hidden diversity.

Karavellia Martynov and Korshunova, 2025 ( type species K. krapivini Martynov and Korshunova, 2025). Genus diagnosis is ‘up to at least seven pre-anal ceratal rows, rhinophores smooth, central teeth of radula with compressed cusp, prostate moderately distinct, penis conical, usual penial stylet present, supplementary gland inserts into penis, distal seminal receptaculum present’.

Nihonbranchus Martynov and Korshunova, 2025 [ type species N. horii (Baba, 1960) ]. Genus diagnosis is ‘up to 2–3 pre-anal rows, rhinophores smooth, central teeth of radula with compressed cusp, prostate distinct, penis conical, usual penial stylet present, distinct, supplementary gland inserts into penis, distal seminal receptaculum present’.

Eubranchulus Martynov and Korshunova, 2025 ( type species E. durrelli Martynov and Korshunova, 2025 ), besides the type species, also includes at least Eubranchulus rupium (MØller, 1842) (= ‘ Galvina ’ olivacea O’Donoghue, 1922 ) and Eubranchulus novik (Grishina et al., 2024). Genus diagnosis is ‘two pre-anal ceratal rows commonly, rhinophores smooth, central teeth of radula with compressed cusp, prostate not distinct, penis conical, usual penial stylet present, indistinct, supplementary gland inserts into penis, distal seminal receptaculum present’.

In Toso et al. (2024: 9), two species, Amphorina andra Korshunova et al., 2020 and Amphorina viriola Korshunova et al., 2020 , were confused and unsustainably synonymized under the name A. viriola Korshunova et al., 2020 without any analysis. This action was erroneous and completely inconsistent with the original descriptions of A. andra ( Korshunova et al. 2020b: 30–4) and A. viriola ( Korshunova et al. 2020b: 27–30) . Here both original names are restored according to Korshunova et al. (2020b). Amphorina viriola Korshunova et al., 2020 is reinstated here with restored diagnosis as: ‘Body up to around 12 mm; large dorsal pigment spots, if present, yellow orange, dull; in specimens with yellow-orange spots on body and cerata there is never any yellow-orange pigment spot or stripe on the tail, but there might be a median whitish line or broken line on the tail; completely pale specimens lack tail spot; light pinkish subapical ring on cerata present; absence of white punctuated line on external edge of foot; cerata commonly moderate in width without distinctly attenuated apices; digestive gland in cerata relatively broad without distinct short branches; up to four anterior rows of cerata; radular formula 31–47 × 1.1.1, copulative stylet relatively long and almost straight, at the top, receptaculum seminis pear-shaped with short distinct stalk between reservoir and long base’. Amphorina andra Korshunova et al., 2020 is reinstated here with restored diagnosis as: ‘Body up to at least 20 mm; large dorsal pigment spots, if present, bright yellow-orange or reddish orange; in specimens with yellow-orange or reddish spots on dorsal side and cerata, there is never any yellow-orange spot or stripe on the tail, but there could be a whitish median line on the tail; completely pale specimens lack tail stripe or spot; light pinkish subapical ring on cerata absent; absence of a punctuated white line or row of dots on the edge of foot; cerata commonly moderate in width without distinctly attenuated apices; digestive gland in cerata relatively broad without distinct short branches; up to four anterior rows of cerata; radular formula 30–37 × 1.1.1, copulative stylet very short and conical, receptaculum seminis subcircular with long distinct stalk between reservoir and rapidly widening base’.

First, we must highlight the correct spelling of the species A. andra Korshunova et al., 2020 [not to be confused with the erroneously renamed ‘ A. viriola ’ in Tosa et al. (2024) not Korshunova et al. (2020b)] and reinstate its original status. Amphorina andra differs from its sister-species, A. farrani , by molecular phylogenetic data and fine-scale morphological patterns ( Korshunova et al. 2020b). Amphorina andra inhabits full-saline environments, whereas A. viriola inhabits predominantly specific semi-brackish environments under the influence of the Baltic Sea, and displays significant differences in coloration from A. andra . Amphorina viriola Korshunova et al., 2020 (original status reinstated) was described in significant details to specifically highlight the fine-scale epigenetic differences of particular groups of Amphorina specimens. Therefore, any potential variations from distantly placed geographic regions ( Toso et al. 2024), and, moreover, any reference to the absence of distinct genetic distances between A. andra and A. viriola , are completely irrelevant for this case and cannot be the basis for justification of synonymy of A. andra and A. viriola . In this respect, it must be noted that ICZN (1999) does not regulate or set any rules on whether one applies morphological or molecular phylogenetics in taxonomic work, and does not set any restriction for the method of separation of species if, for example, they only differ by epigenetic-driven differences in a particular region, as in the case of A. andra and A. viriola , and therefore the original status of A. viriola is restored here.

Thus, the family Eubranchidae , uniquely among members of the superfamily Fionoidea , represents a complete multilevel spectrum, from possessing an ancestral triserial radula (see above justification of its ancestral status for a majority of the suborder Aeolidacea ), which is invariably present without exception in all of the taxa of the family Eubranchidae , to displaying an astonishing genus-level diversity in external features and reproductive systems. This diversity is currently obscured (WoRMS 2024) under the incorrectly applied ‘ Eubranchus ’ name, and further, missing both distinct genetic and soft epigenetic differences at the ‘species’ level, which would reveal true evolutionary patterns. Therefore, the major Eubranchidae diversity patterns are carefully revisited here, and the previously justified fine-scale taxonomy ( Martynov 1998, 2005, Korshunova et al. 2020b), both at the genus and ‘species’ levels, is reinstated.