Adriatic species of Schizomavella (Bryozoa: Cheilostomata)

ABSTRACT In this paper, material belonging to the genus Schizomavella, collected along the Croatian coast of the Adriatic Sea, is revised. Nine species were identified, including five species new to science: S. cornuta, S. halimedae, S. linearis, S. mamillata, S. adriatica sp. nov.,S. mystacea sp. nov., S. rosae sp. nov., S. stanislavi sp. nov. and S. tubulata sp. nov. Previous records of Schizomavella from the Adriatic are also discussed. The checklist of Adriatic Schizomavella species is updated to 11 species; a further two species are doubtful owing to wrong previous identifications. The presence of a calcified ‘hood’ covering the opesia of the suboral avicularium is described and its function is discussed. The morphological diversity of ovicells within the genus Schizomavella is compiled and discussed. http://zoobank.org/urn:lsid:zoobank.org:pub:987D8AE0-1E02-430D-9AB5-50B77BEAF52E


Remarks
The Adriatic material studied in the present work fully matches the redescription of Schizomavella cornuta by Hayward and McKinney (2002). Schizomavella cuspidata (Hincks 1880) was for a long time confused with S. auriculata, until the redescription of both species by Hayward and Thorpe (1995). Many varieties of S. cuspidata were described from the western Mediterranean (see e.g. Calvet 1927;Canu and Bassler 1930;Gautier 1962), which have been considered by Reverter-Gil and Fernández-Pulpeiro (1996) as belonging to morphotypes of this species without taxonomic relevance. Later, Hayward and McKinney (2002) revised samples from the Heller collection, and found that the material of Lepralia cornuta was identical to S. cuspidata s. l., and consequently they renamed it as Schizomavella cornuta; S. cuspidata is a junior synonym. However, it is not definitively clear if S. auriculata var. cuspidata s. s. and L. cornuta are actually conspecific. We have examined samples labelled as S. cuspidata from both the Atlantic and Mediterranean, including the lectotype of S. cuspidata (see Figure 3). As stated by previous authors, the frequency of enlarged avicularia and the degree of development of the suboral umbo are rather variable in different colonies; the width of the sinus is also variable. Hincks (1880) characterised his var. cuspidata (from Guernsey, English Channel) as having a conspicuous spike-like process on the front of the ovicell, and a very prominent avicularium. The lectotype specimen has a wide sinus with small condyles; only a few avicularia are enlarged, most of the time they are small, steeply inclined or even perpendicular to the frontal plane ( Figure 3A, B). Notwithstanding, in other Atlantic material (e.g. NHMUK 1897.5.1.723, from Guernsey, Figure 3C) the condyles appear larger, the proximolateral corners of the orifice are more marked and the avicularia are more frequently enlarged; all of these characters seem to be more similar to the Mediterranean material of S. cuspidata.
On the other hand, Waters (1878) described Lepralia auriculata var. leontiniensis from the Pliocene of Bruccoli (Sicily) which Reverter-Gil and Fernández-Pulpeiro (1996) considered to be a senior synonym of S. cuspidata. However, the identity of Waters' variety is unclear. Original material could not be checked, while the original figure (Waters 1878, fig. 5) seems to show a different-shaped orifice, and avicularia somewhat triangular in shape and frequently detached from the orifice, thus resembling Stephanotheca arrogata (Waters, 1879) (see Reverter-Gil et al. 2012). However, the record of var. leontiniensis made by Waters (1879) from Taranto Bay, Naples, does seem to correspond to S. cornuta. Original material no longer exists in the Waters collection, but we have examined two specimens (MM 2857, MM 3024, Figure 3D) that came from the same locality, first assigned to Schizoporella auriculata var. ornata but later changed to var. leontiniensis. Years later, Hincks (1886) defined Schizoporella auriculata var. spathulata from the Adriatic as having an enlarged suboral avicularium. Having re-examined the type material ( Figure 3E), we believe that this variety is conspecific with S. cornuta.
The 'umbonate' form of S. cuspidata seems to be absent from the Mediterranean. Previous records of similar material from this area (Gautier 1962, as S. auriculata var. cuspidata;revised by Reverter-Gil and Fernández-Pulpeiro 1996), actually belong to a different species (see S. mystacea sp. nov. below). A thorough revision of material assigned to S. cuspidata, and related species from the Atlanto-Mediterranean region will be necessary to ascertain the conspecificity of S. auriculata var. cuspidata and L. cornuta, if at all, but this is outside the constraints of the present paper. Therefore, we provisionally accept the synonymy proposed by Hayward and McKinney (2002). Hayward and McKinney (2002) stated that in S. cornuta, both small and enlarged avicularia show a distinct Y-shaped foramen, but this shape is actually also present in other species (see below Schizomavella mystacea sp. nov. and Schizomavella rosae sp. nov.). According to Hayward and Thorpe (1995) and Hayward and Ryland (1999) this distinct foramen shape is only present in enlarged avicularia.
Schizomavella cornuta seems to be common on the northwest European coast. Its distribution in the Mediterranean, where it has been often reported as S. auriculata (see Reverter-Gil and Fernández-Pulpeiro 1996), is imprecisely known owing to the frequency of its confusion with other species. That being said, S. cornuta must be common in the western Mediterranean as it was the most common species among the Adriatic material studied here.

Diagnosis
Primary orifice short, with very conspicuous proximal notches, surrounded by an even gymnocystal rim including the distalmost part of the suboral avicularium; three to five oral spines; subrectangular suboral avicularium, proximally directed, sometimes slightly laterally displaced, with a distalmost calcified 'hood'; palatal foramen Y-shaped; nonovicellate zooids with a projecting arch distal to the orifice, formed by the secondary calcification of the distal/succeeding zooid(s), rarely developing a small avicularium; ovicell imperforate, with a central umbo and three large, elongate pores just above the orifice; ovicellate zooid developing lateral lapets.

Etymology
From Latin mystax (= moustache), alluding to the shape of the proximolateral notches in the primary orifice.

Description
Colony encrusting, unilaminar to multilaminar, developing as small, irregular crusts. Autozooids rectangular or irregularly polygonal, in radial series, separated by fine, raised sutures; frontal shield nodular, irregularly perforated by 10-20 pores, plus a row of conspicuous areolar pores. A smaller obvious pore either side of the suboral avicularium, placed just below the rim of the primary orifice, sometimes partially concealed by secondary calcification. In young, marginal autozooids, the frontal wall is slightly convex, distally raised to form a slight suboral umbo with which the avicularian cystid is associated; in older, more heavily calcified, zooids, the frontal wall thickens until it reaches the same level as the top of the peristome, so flattening the zooid's appearance.
Primary orifice deeply immersed, the poster tilted basalwards in well-calcified zooids; horseshoe-shaped, much wider than long, its greatest width at midlength; median sinus small, U-shaped, occupying one third of the proximal border. Lateral borders of the primary orifice distinctly curved inwards, describing deep, sharp notches in the proximal corners, accentuated by the edges of the thick, broad, triangular condyles which extend medially beyond the bounds of the sinus. Primary orifice surrounded by a smooth, fine and slightly raised rim, which encompasses the distal end of the suboral avicularium. Nonovicellate zooids with a projecting arch distal to the orifice, formed by the secondary calcification of the succeeding zooid, rarely developing a small avicularium ( Figure 5A). Three to five (usually four) oral spines, as long as an autozooid, but frequently broken. Zooid lateral walls with small uniporous septula, placed in rows near the basal wall. Avicularium median suboral, just proximal to the sinus and sometimes concealing it, or occasionally slightly displaced laterally; elongate, subrectangular, proximally directed, inclined at an angle or even perpendicular to the plane of the orifice; edge of the rostrum sometimes finely denticulated; complete crossbar with a median columella. The distalmost part of the avicularium (the opesia) is placed within the rim of the primary orifice, and covered by a raised, calcified, opesial 'hood'. Rostrum with Y-shaped palatal foramen, with several sharp denticles on the inner edges.
Ovicell not closed by the zooidal operculum, recumbent on succeeding autozooid, covered by a nodular, imperforate secondary calcification, developing a median prominence, sometimes very acute and pronounced. A small, proximal area, immediately above the aperture, with three (rarely four) rounded or elongate pseudopores, frequently two proximal, horizontally orientated, and one medial, smaller, vertically orientated. Ovicellate zooids developing lateral lappets, extending to the ooecium and continuous with the ovicell outer calcification.

Remarks
The set of characters of Schizomavella mystacea sp. nov. outlined above clearly differentiates this species from the others described in herein. Schizomavella mystacea sp. nov. was first recorded by Gautier (1962) as S. auriculata var. cuspidata and by Reverter-Gil and Fernández-Pulpeiro (1996, in part) as S. cuspidata. As stated above, S. cuspidata was considered a junior synonym of S. cornuta by Hayward and McKinney (2002), who redescribed the species and designated a neotype, a conclusion with which we tentatively agree. Schizomavella mystacea sp. nov., however, differs from the redescription of S. cornuta as well as from the lectotype of S. cuspidata mainly in terms of its primary orifice morphology. The primary orifice in S. mystacea sp. nov. is much wider than long, with extremely marked lateral notches; the larger condyles, extending beyond the edges of the sinus; the development of a smooth, thin, even rim around the primary orifice, including the distalmost part of the avicularium; the projecting proximal arch of autozooids distal to the orifice of infertile zooids; the avicularium variable in length, but never enlarged or spatulate, and with its distalmost portion (opesia) covered by a calcified 'hood'; finally, in well-calcified, ovicellate zooids, the rim of the orifice produces two well-developed lateral lappets ( Figure 5G). This final character is also seen in the figures by Hayward and McKinney (2002, fig. 4D), Zabala (1986, pl. 12, fig. D), and Zabala and Maluquer (1988, pl. 16, fig. B), but as the primary orifice is not clearly seen in these illustrations we cannot be sure about the identification of the material. Finally, the number of pores in the frontal shield is higher in S. mystacea sp. nov. than in S. cornuta, while its ovicell has three large, elongate pores, against c. five in S. cornuta.
The Atlantic species S. auriculata also often has lateral lappets in ovicellate zooids, though these are less developed than in S. mystacea sp. nov., but it differs in several other characters, most notably the shape of the primary orifice and the orificial condyles.
The Mediterranean species Schizomavella gautieri Reverter-Gil and Fernández-Pulpeiro, 1997 also develops a stout suboral umbo and a very conspicuous rim around the orifice, but this species differs from S. mystacea sp. nov. mainly in the shape of the primary orifice (subquadrate, with distinct shoulders on each side of the sinus), smaller condyles and the absence of proximolateral orificial notches.
In six of the seven S. mystacea sp. nov. colonies studied, most of the zooids are ovicellate. Only in specimen MNHN 11218, a multilaminar colony with zooids chaotically arranged, are ovicells infrequent. In other species of the genus, ovicellate zooids are, in general, more scarcely produced. Non-ovicellate zooids of S. mystacea sp. nov. have a projecting arch distal to the orifice, formed by the secondary calcification of the succeeding zooid(s) ( Figures 4B, 5A). In material from the Algarve, this projection may rarely develop a small, inner avicularium ( Figure 5A). This morphology has not been seen in any other European species of Schizomavella.
Besides the Adriatic material, we have identified two colonies of S. mystacea sp. nov. in the Gautier Collection collected from Marseille, a dead colony collected in the Cassidaigne Canyon at 230 m depth, and a colony collected in the Algarve (S. Portugal). Therefore, S. mystacea sp. nov. is present in the western Mediterranean, extending from the Adriatic Sea to the Gulf of Cadiz. It thus follows that S. mystacea sp. nov. may have been recorded in this area previously, perhaps as S. auriculata, with which it has previously been confused (as S. cuspidata; see Hayward and Thorpe 1995;Reverter-Gil and Fernández-Pulpeiro 1996), or as any of their varieties.

Diagnosis
Primary orifice rounded trapezoidal, its greatest width across the distal quarter; sinus small, quadrate or U-shaped, occupying one quarter of the proximal border; condyles small, smooth, distal edges of which are angled towards the edges of the sinus. Orifice surrounded by a fine, even rim, describing deep, rounded notches in the proximolateral corners. Three to five long, stout, hollow oral spines; avicularium median suboral, small, close to proximal orifice rim and continuous with it.

Etymology
Named in honour of the senior author's mother, Rosa Mª Gil Carnicer.

Description
Colony encrusting, unilaminar, developing as a small irregular crust.
Autozooids in radial series, separated by fine, raised sutures; rectangular to polygonal, slightly convex. Frontal shield nodular, irregularly perforated by five to 11 rounded pores, plus a row of especially conspicuous areolar pores. A smaller pore (occasionally two) on each side of the distal portion of the suboral avicularium.
Primary orifice rounded trapezoidal, slightly wider than long, its greatest width across the distal quarter; sinus small, quadrate or U-shaped, occupying one quarter of the proximal border; condyles small, smooth, distal edges angled towards the edges of the sinus. Orifice surrounded by a fine, even rim, producing deep rounded notches in the proximal corners.
Three to five (usually four) long, stout, hollow, oral spines, frequently broken.  Avicularium median suboral; small, monomorphic; slightly inclined to frontal shield on a small prominence; close to proximal orifice rim and continuous with it; rostrum subrectangular, proximally directed; crossbar complete with a small median columella; the edges of the foramen in the rostral palate with several sharp denticles on the inner edges.
Kenozooids rare; small, irregularly shaped, occupying spaces between autozooids, provided with an avicularium, similar in size, shape, and orientation to those of the autozooids.
Zooid lateral walls with small uniporous septula, placed in rows near the basal wall.

Remarks
Schizomavella rosae sp. nov. shows close similarities with two other species of the genus: the Atlantic Schizomavella hondti Reverter-Gil and Fernández-Pulpeiro, 1996 and the Mediterranean Schizomavella triangularis Reverter-Gil and Fernández-Pulpeiro, 1997. Schizomavella hondti has an orifice with a much narrower proximal border and a smaller sinus; the condyles are larger, rhomboidal, with faceted surfaces. The suboral avicularium in this species is smaller, oval and lacks a palate; moreover, although it is placed near the sinus it remains unattached to the orificial rim.
Schizomavella triangularis also possesses a suboral avicularium connected with the orificial rim, but it can be distinguished from S. rosae sp. nov. by its roughly triangular primary orifice with its denticulate internal borders, the large condyles extending beyond the edges of the sinus, and the frontal shield perforated by numerous pores.
Schizomavella rosae sp. nov. also shows superficial similarities to S. cornuta, such as the shape of the orifice and their proximal notches, and the Y-shaped avicularian foramen, which was considered by Hayward and McKinney (2002) as characteristic of the latter species. However, S. cornuta produces zooidal orifices that are much wider than long, widest midway, the sinus is wider, and the condyles are oval, with lateral notches and rough distal edge. Finally, S. cornuta produces polymorphic avicularia, which can either be orientated perpendicular to the frontal wall on a suboral umbo or enlarged spatulate, proximally directed and recumbent of the frontal shield, although both are directly in contact with the sinus of the orifice.

Remarks
The Adriatic material studied in the present work fully matches the redescription of S. halimedae made by López de la Cuadra and García-Gómez (2001) who clearly separated it from the very similar species S. discoidea (Busk, 1859). Lepralia discoidea was originally described from Madeira, but it is thought that other records from the European continental shelf may correspond to at least one new undescribed species. López de la Cuadra and García-Gómez (2001) also stated that many previous records of S. discoidea in the Mediterranean are probably referable to S. halimedae, a species restricted to the inner Mediterranean. However, S. discoidea s. l. seems to be also present at least in the western Mediterranean, as well as in the Atlantic, so its presence in the Adriatic is not impossible.

Diagnosis
Primary orifice rounded, with distal and lateral edges forming a continuous curve; no proximolateral notches; surrounded laterally and proximally by a fine, raised peristome, containing the suboral avicularium; sinus small, U-shaped, occupying one third of the proximal border; condyles broad and thick; four or five stout, hollow, oral spines; avicularium almost perpendicular to frontal plane, enlarged and slightly spatulate; basal (distalmost) portion of the avicularium (opesia) covered by a raised, calcified opesial 'hood'; ovicell globose, its distal half covered by a thick, imperforate cover of secondary calcification, and a crescent-shaped proximal area of exposed ectooecium perforated by several rounded pseudopores; ovicellate zooids with a tubular peristome, totally incorporating the avicularian cystid.

Etymology
Alluding to the tubular peristome developed in ovicellate zooids.

Description
Colony encrusting, unilaminar, developing as a small subcircular crust. Autozooids oval or irregularly polygonal, separated by fine sutures, in linear series or losing orientation (perhaps evidence of partial overgrowth or the start of a multilaminar growth).
Frontal shield slightly nodular, convex, distally raised to form a small suboral prominence; perforated by 12-18 rounded pores, plus a row of elongated areolar pores; a smaller pore (occasionally two) on each side of the suboral avicularium.
Primary orifice rounded, slightly wider than long; distal and lateral edges forming a continuous curve; sinus small, U-shaped, occupying one third of the proximal border; condyles broad and thick, emphasising the sinus and reaching the corners of the orifice (i.e. there are no proximolateral notches). Five (sometimes four) stout, hollow, oral spines, articulated, forming an arch around the distal half of the orifice.
The primary orifice is surrounded laterally and proximally by a fine, raised peristome, starting from the first pair of spines and containing the suboral avicularium. The avicularium is almost perpendicular to the frontal plane, enlarged and slightly spatulate, with finely denticulate rostrum; crossbar complete with a thick, triangular columella; basal (distalmost) portion of the avicularium (opesia) covered by a raised, calcified opesial 'hood'; palatal foramen Y-shaped with granular inner edges.
Zooid lateral walls with small uniporous septula, placed in rows near the basal wall.
Ovicell acleithral, globose, prominent, its distal half covered by a thick, imperforate cover of secondary calcification; a crescent-shaped proximal area of exposed ectooecium perforated by several rounded pseudopores, the two in the proximal (outermost) corners clearly enlarged and elongate. Ovicellate zooids with a well-developed peristome, obscuring the primary orifice; the distal portion partially covering the ovicell, the proximal portion enclosing and incorporating the avicularium, forming a tall tube.

Remarks
Shizomavella tubulata sp. nov. shows superficial similarities with S. halimedae and S. discoidea s. l. in the general shape of the primary orifice, in the number and shape of the oral spines, and in the development of a tubular peristome in ovicellate zooids; the ovicell morphology in S. tubulata sp. nov. is also very similar to that of S. halimedae. However, S. tubulata sp. nov. differs from both species due to its wider sinus, its larger condyles, the fine peristome in non-ovicellate zooids that encloses the median suboral avicularium (that is not as enlarged and spatulate as that in S. halimedae) with its Y-shaped rostral foramen, and a calcified opesial 'hood'. Finally, S. tubulata sp. nov. is characterised by the tall tubular peristome in ovicellate zooids which totally encloses the avicularium; in S. halimedae the peristome is interrupted proximally, in other words not enclosing the avicularium, while in S. discoidea s. l. it is interrupted distally, on the ovicell. Schizomavella tubulata sp. nov. produces a fine peristome which includes the suboral avicularium, with a distal calcified opesial 'hood', as described in S. mystacea sp. nov. (see above), but this species differs from S. tubulata sp. nov. in, amongst other characters, its extremely short zooidal orifice. Schizomavella stanislavi sp. nov. (Figure 9; Table 7 Other material examined Schizomavella hirsuta (Calvet, 1927) n. comb. MOM 421601: Holotype of Schizoporella auriculata var. hirsuta Calvet, 1927 (by monotypy).

Diagnosis
Primary orifice rounded, without proximolateral notches; sinus small, quadrate or U-shaped, occupying one third of the proximal border; condyles small, inconspicuous; primary orifice surrounded by an even rim, including the suboral avicularium; three or four long and thick distal spines; avicularium median suboral, subrectangular; distalmost part of the avicularium (the opesia) covered by a raised, calcified opesial 'hood'; rostral palate foramen triangular; frontal wall with minute processes between the pseudopores; ovicell prominent, rapidly covered by secondary calcification, becoming similar in appearance to the frontal shieldthat is, covered in pores and minute processes.

Etymology
Named in honour of the senior author's father, Estanislao Reverter Sequeiros.

Description
Colony encrusting, unilaminar, developing small, irregular, subcircular crusts. Autozooids rectangular to polygonal, in linear series, separated by fine ridges. Frontal shield nodular, evenly perforated by round pseudopores, plus a row of conspicuous areolar pores; a single smaller pore on each side of the suboral avicularium. In young zooids the frontal wall is convex; in older zooids the frontal wall calcification increases until it reaches the same level as the top of the peristome and produces minute pointed processes between the pores, while the pores themselves become deeply immersed and larger.
Primary orifice rounded or somewhat quadrate, deeply immersed in well-calcified zooids; slightly wider than long; small median sinus, quadrate or U-shaped, occupying one third of the proximal border, concealed by the suboral avicularium, often difficult to see in frontal view; condyles small, inconspicuous, with a smooth, straight distal edge and rounded end, just reaching the edge of the sinus; proximolateral notches absent or poorly marked. Primary orifice surrounded proximally and laterally by a fine, even rim, into which the suboral avicularium is incorporated. Three or four, or even five, distal spines, hollow, long and thick, but frequently broken; their bases are present underneath the ovicell along the distal orifice margin.
Zooid lateral walls with small, uniporous septula, placed in rows near the basal wall. Avicularium median, suboral, monomorphic, placed within the oral rim and nearly perpendicular to the frontal wall; rostrum subrectangular, directed frontally; complete crossbar with a small, median columella; distalmost part of the avicularium (the opesia) covered by a raised calcified opesial 'hood' which may conceal the orificial sinus; palatal foramen triangular, with inner edges slightly denticulate.
Ovicell acleithral, prominent, recumbent on the frontal wall of the succeeding zooid, broader than long; surface with several scattered circular ectooecial pseudopores, smaller than those in the frontal shield; rapidly covered by secondary calcification, becoming similar to the frontal shield, with small processes between scattered pores, but with a small, smooth, vertical area, just distal to the primary orifice.

Remarks
Some European species of Schizomavella also develop a frontal wall covered by pointed processes, especially around the orifice, for instance: the Atlantic species S. sarniensis Hayward and Thorpe, 1995 and S. teresae, the Atlanto-Mediterranean S. grandiporosa Canu and Bassler, 1925 and the Mediterranean Schizoporella auriculata var. hirsuta Calvet, 1927. Schizomavella sarniensis differs from S. stanislavi sp. nov. in having larger condyles with finely toothed distal edges, a less-developed orificial rim that just touches the suboral avicularium but does not include it, and the larger avicularium without the distal calcified 'hood', which is occasionally enlarged and occupying most of the frontal wall.
Schizomavella teresae differs from S. stanislavi sp. nov. in the shape of the primary orifice, with its almost drop-shaped outline and narrower proximal edge, its broad condyles with pointed corners projecting above the sinus, and its smaller avicularium, without a palate or distal calcified 'hood', that does not project above the peristome.
Schizomavella grandiporosa, as recently redescribed by Souto et al. (2013), differs from S. stanislavi sp. nov. in having a wider and shallower sinus, a less developed orificial rim, which is in contact with the shorter suboral avicularium but not engulfing it, and the distinct larger pores in the frontal wall. Calvet (1927) described Schizoporella auriculata var. hirsuta based on a single colony from Monaco, characterising it based on the development of frontal calcification covered by minute pointed processes on both the autozooids and ovicells. We have revised the holotype of this variety (MOM 421601, Figure 10), and believe it should be upgraded to species level as Schizomavella hirsuta. This species differs from S. stanislavi sp. nov. mainly due to the primary orifice, with its straight lateral borders, the distinct shoulders on each side of the sinus and the larger condyles producing lateral notches (Figure 10 B, C). Other differences include a suboral avicularium which projects well above the frontal surface on a somewhat quadrangular prominence (the distal calcified 'hood' of the avicularium is frequently asymmetrical; see Figure 10D), and the heavily developed process-covered peristome that conceals the primary orifice; in S. stanislavi sp. nov. the primary orifice is always clearly seen.
However, it must be noted that all four of these species may appear quite similar when studied using only binocular microscope; a clear differentiation is only made possible by using an SEM on thoroughly cleaned material.
The primary orifices of S. stanislavi sp. nov. and S. tubulata sp. nov. are similar in the absence of proximal notches but S. tubulata sp. nov. differs in the stout oral spines forming an arch, and its stouter condyles. Moreover, the peristome in S. tubulata sp. nov. is more highly developed, forming a tube, and its ovicell is also different from that of S. stanislavi sp. nov. (Hassall, 1841) ( Figure 11; Tables 8-10 Figure 11F).

Remarks
Schizomavella linearis was redescribed by Hayward and Thorpe (1995), who selected a neotype. Its supposed distribution ranges from west Norway and Faroe Isles in the north to the western Mediterranean, and into the Adriatic, in the south. Having examined several specimens from different sampling sites, we have noted some morphological variations within them, and as a result we have outlined these variations in the three 'forms' noted below.
Material coming from Sts. 7, 8 and 11 (Figure 11 A, B) fits the redescription of S. linearis by Hayward and Thorpe (1995) (see also Hayward and Ryland 1999;Hayward and McKinney 2002), and is here named the 'typical' morphotype.
Material coming from Sts. 3, 4, 12, 14, and 16 develops a single suboral avicularium on a large, conical umbo proximal to the sinus, and is here named the 'hastiformis'morphotype. The extent of this umbo, actually the avicularian cystid, varies in morphology from a low, massive cone, to a long, finger-like projection ( Figure 11C-E). The material cited here was previously reported as S. hastata by . The true identity of S. hastata was uncertain until its redescription by Hayward and Thorpe (1995). Earlier authors (and even some later ones) based their identifications on the presence of the single, suboral avicularium on a large umbo; however, the character that best defines S. hastata is the shape of the orifice, it being rounded, with a broad, shallow sinus occupying its entire proximal border, and flanked by very small condyles. On the contrary, the material here studied has a primary orifice quite similar to the typical material of S. linearis, so we think that it does not correspond to S. hastata. A similar morphology can be also seen in material identified as S. hastata from La Atunara (Strait of Gibraltar; López de la Cuadra 1991: pl. 26, fig. F), Marseille (photos sent by J.G. Harmelin), Algeria (MNHN 11239, Gautier Coll.) and the Balearics (MNHN 11222, Gautier Coll.). The discovery of specimens from the north of the Bay of Biscay with a single suboral avicularium led Hayward and Ryland (1978) to describe the species Schizomavella hastiformis, which nevertheless they later considered to be a morphotype of S. linearis (see Hayward and Ryland 1999). Something similar may occur with Mediterranean material previously identified as S. hastata.
Finally, material coming from Sts 1, 2, 6 and 7, here named the 'pseudolinearis' morphotype, has a larger primary orifice that is longer than wide, and the avicularia are larger than in the typical form of the species, and frequently distally orientated, instead of medially ( Figure 11F). Similar colonies have been observed off Avilés (northwest Spain) at 144 m depth, and off Brittany at 255 m depth (unpublished data). We are unable to discount the possibility that these differences are within the range of variation of the species, and we therefore tentatively include these specimens in S. linearis.
These three 'morphotypes' do not seem to follow any pattern of distribution along the Adriatic coast, and no particular abiotic pattern was observed that can explain the morphological variations. Taking into account the high degree of variation observed in material of S. linearis, not only in the Adriatic but also in the western Mediterranean, we believe that it is not impossible that we were dealing with several similar species. To corroborate this assumption, however, it will be necessary to undertake a thorough revision of Atlanto-Mediterranean specimens previously identified as S. linearis or any of its 'varieties', as well as S. hastata, using SEM observations, as well as utilising newly collected material for molecular work, and previous literature. This huge task is outside the scope of the present paper.

Remarks
The Adriatic material studied here fully matches the description and illustrations of Schizomavella mamillata given by Hayward and McKinney (2002), who selected a lectotype for the species. It is noticeable that the autozooidal operculum of the material studied here has a median prominence running proximo-distally, which can be indicative of a strengthening structure, within or below the operculum. This structure has not been observed in other species treated here. Schizomavella mamillata seems to be common throughout the shallow waters of the Mediterranean, but it is also present in several localities along the Portuguese coast ). In the Adriatic, it has only been recorded in recent years: in Jabuka Islet , Rovinj (Hayward and McKinney 2002), Lastovo and Prvić Islands , and several other localities along the Croatian coast, between 10 and 45 m depth .
( Figure 13; Table 12) ?  the rimbut frequently missing from large areas of the colony. Ovicell evenly perforated, soon covered by granular secondary calcification.

Etymology
Pertaining to its only known occurrencethat is, only found in the Adriatic Sea.

Description
Colony encrusting, multilaminar, forming large, irregular patches. Autozooids distinct, separated by fine sutures, arranged in radial series in the basal layer, randomly orientated in frontally budded patches. Autozooids in the basal layer rectangular or polygonal in shape, irregularly polygonal in successive layers. Frontal shield flat or slightly convex, uniformly perforated by rounded pores, each in a distinct depression in newly formed individuals. In older zooids, the increase in secondary calcification gives the frontal wall a rough, granular appearance and the pores appear larger ( Figure 13B); in heavily calcified zooids, the medial pores may become occluded, with the exception of the row of marginal pores which become elongate ( Figure 13A). Zooid lateral walls with small uniporous septula, placed in rows near the basal wall.
Primary orifice rounded, nearly as long as wide. Proximal border with a wide, shallow sinus, occupying half of its width, flanked by two shoulders. Condyles smooth, moderately developed, extending from the edges of the sinus to the lateral edges of the orifice and continuous with it.
Suboral avicularium small, oval, slightly longer than wide, quite distant from the primary orifice and unattached to the rim. Proximally directed, with rostrum almost parallel to frontal plane. Crossbar complete, slender, sometimes forming an open angle in the middle but without columella. Opesia transversely oval or subcircular; palatal foramen large, oval, with semielliptical mandible. However, avicularia are frequently missing from large areas of a colony.

Remarks
Schizomavella adriatica sp. nov. was previously reported as Schizomavella rudis (Manzoni, 1869) by Hayward and McKinney (2002) from the Northern Adriatic, by Novosel (2007) from several Croatian localities, and probably by McKinney and Jaklin (2000). Reverter-Gil et al. (2012) recently re-examined many Mediterranean records of S. rudis and as a result described a new species, Stephanotheca watersi. However, the record of Hayward and McKinney (2002) has been considered to correspond to another, as yet, undescribed species of Schizomavella. Among other differences with S. rudis, the zooidal orifice of S. adriatica sp. nov. has a proximal edge with a shallow sinus flanked by two shoulders and smooth condyles, the ovicell is uniformly perforated and is not closed by the operculum and S. adriatica sp. nov. does not exhibit a dimorphic orifice in ovicellate zooids. Schizomavella adriatica sp. nov. shows some similarities with S. sarniensis and S. grandiporosa, as recently redescribed by Souto et al. (2013); for instance, it has the rounded orifice with a shallow, wide sinus, the frontal shield evenly perforated by large pores and the ovicell immersed by secondary calcification. However, the shape of the condyles is different amongst these three species: in S. adriatica sp. nov. they are smooth and continuous with the lateral walls of the orifice; in S. grandiporosa they are smooth, oval, with a lateral notch; and finally in S. sarniensis they are large, finely toothed and also notched. Differences are also seen in the avicularia amongst these three species. In S. adriatica sp. nov. the suboral avicularium is frequently absent from large areas of a colony, but when present it is distant from the primary orificethat is, unattached to the orificial rim, with the rostrum parallel to frontal plane, the crossbar lacking a columella. In the other two species the avicularia are nearly always present, close to the sinus and continuous with the orificial rim, almost perpendicular to the frontal wall, and their crossbars have a well-developed columella. Hincks (1886) reported the presence of S. auriculata in the Adriatic and defined the new variety S. auriculata var. spathulata. It is unclear if he considered all his Adriatic material to belong to this variety, or whether part of it represented the 'typical' morphology. We have examined three samples from the Adriatic in the Hincks collection (NHMUK 1899.5.1.459, NHMUK 1899.5.1.468, NHMUK 1899, originaly labeled as Schizoporella auriculata (later reassigned as Schizomavella sp. indet. by P. Hayward); this material is here assigned to S. adriatica sp. nov.

Other species previously reported from the Adriatic Sea
Schizomavella asymetrica (Calvet, 1927) Hayward and McKinney (2002) upgraded Calvet's (1927) variety asymetrica to species status and reported it for the first time in the Adriatic. Schizomavella asymetrica was originally recorded from Monaco (Calvet 1927) and subsequently reported by Gautier (1962) from Marseille and Tunis.  record corresponds to S. stanislavi sp. nov. (see above). Schizomavella asymetrica was not found during the present study.
Schizomavella auriculata (Hassall, 1842) Several authors have reported S. auriculata from the northern and central Adriatic (see Novosel and Požar-Domac 2001;. However, it must be stated that S. auriculata is an Atlantic species that is absent from the Mediterranean. Its diagnostic characters, previously confused with those of S. cuspidata (= S. cornuta) as outlined above, were established by Hayward and Thorpe (1995). Previous records of S. auriculata in the Mediterranean correspond to various other species of Schizomavella, most often S. cornuta (see Reverter-Gil and Fernández-Pulpeiro 1996); however, the revision of all original material is necessary to validate the identifications.
Schizomavella discoidea (Busk, 1859) This species has previously been reported from the northern and central Adriatic (Hincks 1887;McKinney 2000;McKinney and Jaklin 2000;Zavodnik et al. 2000;. Lepralia discoidea was originally described from Madeira. Its type material was not found at the NHMUK, but we have examined Madeiran material in the Norman Collection. Comparison of this Madeiran material with European continental shelf records (from Great Britain to the Iberian Peninsula in the Atlantic, as well as the Western Mediterranean: see Hayward and Ryland 1999;López de la Cuadra and García-Gómez 2001;Reverter-Gil and Fernández-Pulpeiro 2001), strongly suggests that these records actually correspond to at least one new undescribed species, but a thorough revision of the material will be necessary to elucidate this any further. Notwithstanding, S. discoidea s. l. and S. halimedae (found during the present study) are morphologically similar species that have only recently been clearly separated (see López de la Cuadra and García-Gómez 2001). As stated by these authors, many previous records of S. discoidea s. l. in the Mediterranean are probably referable to S. halimedae, a species restricted to the inner Mediterranean. In fact Hincks (1887, p. 303), who first reported S. discoidea from the Adriatic, stated that 'In specimens from the Adriatic the small avicularium on the front of the cell is sometimes replaced by a long spatulate form. The dependent lateral appendages were not noticed'. This description clearly corresponds to S. halimedae. However, S. discoidea, or another very similar undescribed species, seems to be present at least in the western Mediterranean. Therefore, the presence of S. discoidea in the Adriatic cannot be discounted, but needs further investigation to be confirmed. No colonies of S. discoidea were found during the present study.
Schizomavella gautieri Reverter-Gil and Fernández-Pulpeiro, 1997 This species was originally described from Algeria, and later found in Isola Vulcano (Italy) (J.-G. Harmelin unpublished data: MNHN 20204). It was wrongly reported from the Adriatic by ; her original material has been re-examined and belongs to S. cornuta (see above). Schizomavella gautieri was not found during the present study.
Schizomavella hastata (Hincks, 1862) This species was previously reported in the Adriatic by ), Friedl (1918. However, the specific characters of this species were unclear until its redescription by Hayward and Thorpe (1995). Material documented by  is here reassigned to S. linearis (see above). It should be stressed that the most useful character for distinguishing S. hastata from other species of Schizomavella is the shape of the primary orifice, not the size and position of the suboral avicularium. In a reexamination of reference illustrations and material assigned to S. hastata from the Mediterranean, we have noticed that many orifices actually resemble those of S. linearis rather than S. hastatathat is, they never show the characteristic shallowly concave proximal border seen in S. hastata. In conclusion, S. hastata is probably absent from the Adriatic, and perhaps from the whole Mediterranean, although it was previously considered to be present in the area (Reverter-Gil and Fernández-Pulpeiro 1996;Hayward and Ryland 1999). All previous records will need to be reassessed with reference to their original material to ascertain its true distribution.
Schizomavella rudis (Manzoni, 1869) This species was reported in the Adriatic by McKinney and Jaklin (2000), Hayward and McKinney (2002) and . As stated above, all of these records probably correspond to S. adriatica sp. nov. Hayward and McKinney, 2002 This species, described by Hayward and McKinney (2002) from the northern Adriatic, was later reported by . Unfortunately, her original material has been lost, and so was not included in the present study. Reverter-Gil and Fernández-Pulpeiro, 1997 This species was reported by Chimenz Gusso et al. (2007) from the northern Adriatic, but it was a typographical mistake (Chimenz Gusso pers. comm.); it was deleted from that area in a subsequent paper (Rosso et al. 2010). Schizomavella triangularis was not found during the present study.

Discussion
Among the species described in this paper, three of them have a calcified 'hood' covering the opesia of the suboral avicularium: S. stanislavi sp. nov., S. mystacea sp. nov. and S. tubulata sp. nov. This structure was only recently reported in Schizomavella, present in an unknown species from Madeira ), but it is also present in S. hirsuta from Marseille (see above). This 'hood' is formed when the peristomial rim surrounds the primary orifice and encloses the avicularium, or at least its distal part. In all cases, the avicularium is close to the proximal border of the orifice and almost perpendicular to the frontal plane. The calcified 'hood' seems to protect the opesia of the avicularium when the zooidal operculum opens, working as a door-stop, a function perhaps similar to the lyrula of other ascophorans (see Berning et al. 2014). However, in other Schizomavella species with the avicularium in a similar position, such as S. auriculata, S. cristata (Hincks, 1879) and S. teresae, the 'hood' is absent. Further studies will be necessary to understand the development and function of this structure.
In Schizomavella, different morphologies of ovicell can be found. In the genotype species, S. auriculata, the ovicell is acleithral and the ooecium is covered by an imperforate layer of secondary calcification, but there is a proximal area of exposed pseudoporous ectooecium. This morphology is present in many other species of the genus, with some variation regarding the shape and extension of the pseudoporous area, the number of pores, and their size: S. cornuta, S. halimedae, S. linearis, S. mamillata, S. mystacea sp. nov. and S. tubulata sp. nov., among the species treated in the present paper; S. asymetrica, S. cristata, S. fischeri, S. gautieri, S. hastata, S. hondti, S. triangularis and S. subsolana, among other European species.
In several species previously assigned to Schizomavella, the ovicell is cleithral. In these species the ovicell opening is formed by the concave proximal margin of the ooecium, which extends to the proximolateral corners of the zooidal orifice; the periphery of the ooecium is covered by a thick layer of imperforate secondary calcification, leaving a central rounded area of exposed pseudoporous ectooecium. These species were recently transferred to the lanceoporid genus Stephanotheca (see Reverter-Gil et al. 2012).
In other European Schizomavella species, such as S. discoidea s. l., S. neptuni (Jullien, 1882) and S. noronhai (Norman, 1909), the globular ovicell is also acleithral. In these species the ooecium only has a basal, peripheral rim of imperforate secondary calcification and a large area of exposed pseudoporous ectooecium which may be covered by a thin layer of secondary calcification that does not occlude the pseudopores (Reverter-Gil et al. forthcoming).
Schizomavella stanislavi sp. nov., together with other European species S. grandiporosa, S. hirsuta, S. sarniensis and S. teresae, has a slightly different ovicell morphology: the ooecium is initially smooth, randomly perforated by pseudopores, but later covered by a thick, nodular secondary calcification that does not occlude the pseudopores. However, in these species there is always a small, uncovered vertical portion of the pseudoporous area just distal to the primary orifice (see Reverter-Gil and Fernández-Pulpeiro 1996;Souto et al. 2013;present paper), not too dissimilar to the type species of the genus. In all these species, both the frontal shield and the ovicell become heavily calcified and covered by minute, pointed processes, which makes it difficult to differentiate them without SEM imagery. This perhaps indicates that these species are closely allied, but probably they do not constitute a separate clade. Two of them, S. sarniensis and S. teresae, seem to be present only in the European Atlantic. Schizomavella grandiporosa is present from the northwest of the Iberian Peninsula down to North Morocco and Algeria. Finally, S. hirsuta and S. stanislavi sp. nov. are probably endemic Mediterranean species; both species share the distal calcified 'hood' on the suboral avicularium and perhaps both are derived from a 'S. grandiporosa-like' ancestor. If this is the case, S. mystacea sp. nov. and S. tubulata sp. nov., with their distinct morphologies, perhaps developed the calcified 'hood' independently, making it a functional, analogous struture. Schizomavella mystacea sp. nov. may be derived from a 'S. cornuta-like' ancestor, as this species has a similar ovicell, with S. tubulata sp. nov. deriving from a 'S. halimeade-like' ancestor, both sharing the same ovicell.
Finally, S. adriatica sp. nov. also has an ovicell covered by perforated secondary calcification, but in this case there is no remnant of the pseudoporous ectooecium proximally. Also, the shape of the primary orifice in S. adriatica sp. nov. differs from other species treated here, and its avicularium, when present, is placed far from the orifice. As such, S. adriatica sp. nov. does not have a close morphological relationship with the other species treated in the present paper.

Conclusions: updated list of Adriatic Schizomavella species
As a result of the revision of material in the present work, as well as the available literature, we can conclude that 11 species of Schizomavella are presently known in the Adriatic Sea. The presence of two other species is here considered uncertain (see below).
The most diverse locality in the Adriatic is Jabuka Shoal, where eight Schizomavella species were recorded. Rovinj and Vis (Komiža Bay) both have six species, while Jabuka Islet, Biševo and Korčula and Lastovo Islands have five species each.