Two new species of cheilostomate Bryozoa from Iberian waters

Two new species of cheilostomate bryozoans are described from material preserved in museums: Cradoscrupocellaria severoi sp. nov., from Iberian Mediterranean waters, and Setosella margaritae sp. nov., from shallow waters along the Atlantic coast of Europe. Moreover, the Mediterranean species Setosella cyclopensis Rosso, Di Martino & Gerovasileiou, 2020 is reported in Iberian waters for the fi rst time.


Introduction
Although no complete catalogue is currently available, we consider that the bryozoological fauna of the Iberian Peninsula is one of the best known in European waters, with approximately 540 recent species cited in this region (about 420 along the Atlantic coast and about 340 along the Mediterranean coast; own unpublished compilation). Nonetheless, our overall knowledge remains fragmentary and undoubtedly includes many taxonomic errors, which are gradually being corrected in various studies. The greater ease of access to original reference material and, above all, the use of the electron microscope enable better characterization of species, and makes possible to detect misidentifications and misinterpretations of the species concept as well. Revision of material stored in collections has thus become an essential part of the work of taxonomists, either to redescribe known species or to describe new ones.
The genus Cradoscrupocellaria Vieira, Spencer Jones & Winston, 2013 was recently described after a major revision of species originally ascribed to the genus Scrupocellaria s. lat. Cradoscrupocellaria reptans (Linnaeus, 1758) was redescribed by Vieira & Spencer Jones (2012). On the other hand, Rosso et al. (2020) also very recently redescribed the genus Setosella Hincks, 1877, andSouto et al. (2016) redescribed its type species, S. vulnerata (Busk, 1860). These taxonomic revisions made it necessary to study and correct previous records in order to update the identification of certain specimens. Based on the revision of material preserved in museums, we present here two new species of cheilostomate Bryozoa belonging to the genera Cradoscrupocellaria and Setosella. We also report the Mediterranean species Setosella cyclopensis Rosso, Di Martino & Gerovasileiou, 2020 in Iberian waters for the first time. This extends our knowledge of the bryozoan fauna in Iberian waters and in Europe as a whole.

Institutional abbreviations
Samples deposited in the collections of the following museums have been revised: MHNUSC = Museo de Historia Natural da Universidade de Santiago de Compostela MNCN = Museo Nacional de Ciencias Naturales, Madrid MNHN = Muséum national d'histoire naturelle, Paris NHMUK = Natural History Museum, London The samples were examined with stereomicroscopes and uncoated, clean specimens were selected to photograph with scanning electron microscopes: a Zeiss EVO LS15 (Santiago de Compostela, Spain) and an Inspect S50 (Vienna, Austria and Madrid, Spain), always with a back-scattered electron detector in low variable vacuum mode. Measurements were taken with the software ImageJ® on SEM photographs.

Etymology
This species is dedicated to Severo Ochoa de Albornoz (1905Albornoz ( -1993, Spanish physician and biochemist, winner of the 1959 Nobel Prize in Physiology or Medicine for the discovery of the mechanisms in the biological synthesis of nucleic acids.  (Fig. 3).

Description
Colony erect, up to 1 cm in height, branched, with internodes comprising 5-10 alternating autozooids. Internodes slender, almost straight, angled at axis, with acute bifurcating pattern; chitinous joints passing across proximal gymnocyst of outer (C and D) and inner zooids (F and G) at bifurcation. Autozooids cylindrical with straight sides. Oval opesia occupying two thirds of zooidal length; cryptocyst reduced to narrow rim around opesia. Large scutum inserted at midline of the inner edge of opesia, stout but relatively flat, its tips highly branched, fully developed and completely covering entire opesia. Distal spines short, unbranched; 2 inner and 3 outer spines, with outer distal spines closer to scutum; axial zooid with 5 spines. Small distolateral avicularium frequent and easily visible in frontal view. Monomorphic frontal avicularium often present in outer zooids of each internode; large but variable in size, with elongate rostrum 0.22-0.28 mm long, with fringed edges, proximally upright, with strongly hooked tip; mandible long, hooked distally. Vibracular chamber laterally placed on the basal surface of each zooid, visible in frontal view; chamber of vibraculum almost trapezoidal, with a proximal rhizoidal foramen; setal groove transverse to internode axis, straight, with smooth seta longer than one autozooid. Single axial vibraculum without rhizoidal foramen. Rhizoids tubular and smooth. Ovicells globular, wider than long, with an even, almost straight proximal rim; ectooecium perforated by 10-18 rounded pores; ovicelled zooids with 2 inner and 2 outer spines. Ancestrula unknown.

Remarks
The genus Cradoscrupocellaria has been recently described by Vieira et al. (2013). Nine already existing species were transferred to this genus, and the same authors described 18 new species. Among all these species, only C. gautieri Vieira, Spencer Jones & Winston, 2013 and C. macrorhyncha (Gautier, 1962) show similarities with C. severoi sp. nov., mainly regarding the very stout scutum with truncate tips and the large frontal avicularia. The description of C. gautieri is based on a single sample collected by Gautier (1962) in Algeria. Cradoscrupocellaria macrorhyncha was redescribed by Vieira et al. (2013) based on the only original sample of Gautier deposited at NHMUK, designated as lectotype, and other Mediterranean samples by other authors. However, the MNHN collection includes two other original samples of Gautier from the same station as the lectotype that seem to have been overlooked in the redescription of the species, perhaps because in the museum's catalogue and in the external labels of the samples both are simply marked as "Scrupocellaria reptans": sample MNHN-IB-2008-11319 contains a handwritten label by Gautier stating "Scrupocellaria reptans group? Dry specimen"; being written in English, it is highly probable that this sample together with the lectotype were sent by Gautier to A. Hastings at the NHMUK, and later returned. Sample MNHN-IB-2008-11321 ( Fig. 3) includes a label handwritten by Gautier indicating "Scrupocellaria reptans? grand avic.". Since both samples actually correspond to C. macrorhyncha and were collected by Gautier at the type locality of the species and studied by him, they should be considered as part of the type series and consequently considered as paralectotypes of the species according to Arts 73.2.2 and 74.1.3 of the ICZN Code (ICZN 1999).
Cradoscrupocellaria severoi sp. nov. differs from C. macrorhyncha and C. gautieri by several characters: the frontal avicularium is monomorphic in C. severoi sp. nov. because the small columnar avicularium present in the other species is absent; this frontal avicularium is oriented upwards on the branch, not proximally (Figs 2A-B, 3C). The lateral avicularium is more frequent and more visible ( Fig. 2A-B). In the bifurcations of C. severoi sp. nov., the chitinous joints pass through the gymnocyst of the external zooids (C and D) instead of through their opesiae (Figs 1C, 3A). The ovicell is wider than long in C. severoi sp. nov., usually uniformly perforated by small pseudopores (Figs 1D, 2A-B), whilst in C. macrorhyncha the ovicell is more spherical and usually exhibits an imperforate flat frontal area (Fig. 3B) (in C. gautieri the ovicell is unknown). The autozooids of C. severoi sp. nov. are clearly smaller than those of C. macrorhyncha, but similar in size to those of C. gautieri. Finally, the internodes are short in C. severoi sp. nov., formed by 5-10 autozooids (Fig. 1A-B), as in C. gautieri, but the scutum is flatter and not swollen as in that latter species.  Cradoscrupocellaria severoi sp. nov. is known with certainty only from its type localities at the Columbretes Islands (Mediterranean Spain) between 3 and 42 m depth. Zabala (1986) reported Scrupocellaria macrorhyncha Gautier, 1962 from several Catalonian localities, but these records were later reassigned to S. reptans (Linnaeus, 1758) by Zabala & Maluquer (1988). Vieira et al. (2013), however, stated that these specimens are distinct from C. macrorhyncha and C. reptans, as well as from C. ellisi (Vieira & Spencer Jones, 2012), and that they may belong to a new undescribed species. Figures by Zabala (1986) and Zabala & Maluquer (1988) are quite similar to C. severoi sp. nov., but the description given (actually an almost literal translation of the original description of S. macrorhyncha by Gautier) show several differences from the present description of the new species. A revision of the original material and collection of new material will be necessary to clarify which species was reported from Catalonia, as well as to determine the possible presence of C. severoi sp. nov. at other Iberian localities.

Differential diagnosis
Setosella with small, encrusting colonies. Autozooids small, oval, with opesia D-shaped or irregularly rounded, and two oval to tear-shaped opesiules, positioned close to opesia and directly beside lateral walls of zooid. Small interzooidal vibracula oval, positioned distolateral to each autozooid, always on the right side and often without exceeding distal edge of autozooid, especially in ovicelled ones. Ectooecium with transversely oval membranous window and granular endooecial surface underneath, with a small, central pore. Ancestrula oval, with cryptocyst occupying slightly less than half of the frontal area; opesia semielliptical, with straight or slightly concave proximal border.

Etymology
This species is dedicated to Margarita Salas Falgueras (1938-2019), Spanish scientist, medical researcher, and author in the fields of biochemistry and molecular genetics. She was a disciple of S. Ochoa (see above).

Paratypes
ATLANTIC SPAIN • several small juvenile colonies on shell fragment; same collection data as for holotype; MHNUSC 10121 ( Fig. 4C) • colony on shell fragment; same collection data as for holotype; MHNUSC 10122 ( Fig. 5C-D

Description
Colony encrusting, unilaminar, forming small discoidal patches of alternating autozooids and vibracula. Autozooids irregularly oval, with well-developed smooth gymnocyst proximally that narrows and steepens distally, lateral walls slightly raised, framing an evenly granular cryptocyst that is flat and depressed proximally, gently rising distally to the opesiules to form the proximal border of the opesia. Opesia D-shaped or irregularly rounded, wider than long, distal margin with some blunt, irregularly spaced denticles. Two oval to tear-shaped opesiules (ca 20 μm long by 10 μm-wide), located in distal depressed area of the cryptocyst, positioned close to the opesia (mean 36 μm) and directly beside the lateral walls of zooid, their inner edges sometimes with several sharp denticles; the size of both opesiules unequal, the left one slightly larger. Small interzooidal vibracula oval, positioned distolateral to each autozooid, always on the right side and often without exceeding the distal edge of the autozooid, especially in ovicelled ones. Wide oval opesia, sometimes slightly narrower in the middle; seta long and slender, curved, up to twice length of the autozooid. Communication of zooids via small uniporous septula. Some autozooids and vibracula show evidence of breakage and regeneration associated with the intramural budding (Fig. 5C-D). Ovicells terminal, with a brood cavity immersed within the distal part of the maternal zooid. Kenozooidal ooecium roughly level with the colony surface, forming shallow hood covering distal end of the maternal zooid from which it is budded. Proximal ooecial margin forming the distal part of the zooidal orifice; ectooecium with transversely oval membranous window and granular endooecial surface underneath, with a small, central pore. Ovicellate zooids dimorphic, slightly wider distally, with orifices distinctly broader and campanulate in outline. Distal budding of autozooids and vibracula in ooecium-producing zooid retained. Ancestrula oval, with cryptocyst relatively smooth, occupying slightly less than half of the frontal area; opesia semi elliptical, with straight or slightly concave proximal border. Astogenesis beginning with one distal and two lateral autozooids; later zooids more irregularly arranged. The ancestrula also buds the two typical caudate vibracula of the genus: one short caudate, budded mid-laterally on the left side, and the other long caudate, budded distally, and sometimes curved to the right. On one occasion, however, this pattern was reversed, with the short vibraculum budded on the right side (Fig. 4C).
Setosella margaritae sp. nov. differs from S. vulnerata (see redescription by Souto et al. 2016) as well as from S. cyclopensis, quite a similar species, by several characters: the vibracula of S. margaritae sp. nov. are much smaller, about half the size, and characteristically shifted laterally on the right side, often without exceeding the distal end of the autozooid (especially in ovicelled zooids), instead being distal or only slightly distolateral. The autozooids are oval and clearly smaller in S. margaritae sp. nov. The opesiules are shorter, oval to tear-shaped, instead slit-like or elongated; moreover, the opesiules are located closer to the opesia and directly beside the lateral walls of the zooid, instead of away from the opesia and the lateral walls. The window of the ectooecium in S. margaritae sp. nov. is transversally oval, whilst in S. vulnerata and S. cyclopensis it is roughly circular, much smaller in the former species, much larger in the latter. Finally, the colonies of S. margaritae sp. nov. are very small, encrusting mainly shell Fig. 6. A colony of Setosella vulnerata (Busk, 1860) -Gil et al. 2012) differ by their circular opesiules, four or even up to five in the latter two species. Finally, S. spiralis differs by the much larger autozooids and vibracula, with opesiules located further away from the opesia, and by colonies with spirally arranged zooids typically in a single rightcoiled row.
In the past, we have considered our own material to be similar to typical Setosella vulnerata.  Hayward & Ryland (1998) fit the present description of S. margaritae sp. nov., at least the shallow-water material referred to there. Accordingly, the species is quite possibly distributed in shallow waters along the Atlantic coast of Europe, from the North Sea to at least the NW of the Iberian Peninsula. Its occurrence in other areas should be confirmed by reviewing previous citations of S. vulnerata. In a previous paper (Reverter-Gil et al. 2012), we already suggested that previous records of S. vulnerata might correspond to several different species.
There are a few more previous records of Setosella vulnerata in Iberian waters: the record from Alboran Island, at 118 m depth, made by Templado et al. (2006) actually corresponds to S. vulnerata (see material examined and Fig. 6) as well as the recent records published by Ramalho et al. (2020) in a nearby area between 95 and 440 m depth. However, the record from Columbretes Islands (Mediterranean Spain), at 80 m depth, made by Templado et al. (2002) actually belongs to S. cyclopensis (see material examined; unfortunately, this sample is currently unavailable for photography). Thus, this is the first Iberian record of this Mediterranean species. We have no further information about the records published by Zabala et al. (1993) from the Blanes Canyon at 180-350 m depth and by Madurell et al. (2013) from Cap de Creus at 104-225 m depth (Catalonia), but based on the given depths these records may belong to S. vulnerata or to S. cyclopensis.
Following Souto et al. (2016) and Rosso et al. (2020), Setosella vulnerata is distributed in the Northeast Atlantic and the Mediterranean, although several Atlantic and Mediterranean occurrences still need to be checked. As already stated by Rosso et al. (2020) it is likely that S. vulnerata is actually restricted to deep habitats from the shelf break and the continental slope.

Discussion
As noted in the Introduction (see above), the bryozoological fauna of the Iberian Peninsula is one of the best known in European waters. Our own unpublished compilation, based on dozens of articles published over the last century and a half, and the revision of hundreds of samples -both our own and those of museum collections -has yielded approximately 540 recent species cited in this region. In comparison, only 556 species have been registered so far in a larger area as studied as the Mediterranean Sea as a whole, where there is an extensive bibliography on bryozoans dating back more than 200 years (Rosso & Di Martino 2016). Moreover, all European waters combined have yielded 945 indigenous species according to Gordon et al. (2019). The description of two new species in a supposedly so wellknown area underlines the continued need for purely taxonomic and faunal works: they are key pillars to develop well-designed and useful biodiversity conservation policies (e.g., Wägele et al. 2011;Higgs 2017;Thomson et al. 2018).