taxonID	type	description	language	source
5947A847E94FFFE13FE9C272FA53FD2D.taxon	description	$ These authors contributed equally to this work † Current address: Centro Piattaforme Tecnologiche, Università di Verona, Piazzale A. Scuro 10, 37134 Verona, Italy [Version of record, published online 31 March 2020; http: // zoobank. org / urn: lsid: zoobank. org: pub: DB 18 D 5 A 7 - 0 E 96 - 458 E-B 3 C 0 - 5 E 93327359 AC] become a reference organism, not only in fields such as evolutionary developmental biology and immunobiology, but also for studying biological processes such as stem cell migration, apoptosis, regeneration and ageing (Sabbadin, 1982; Manni et al., 2007, 2019; Rosengarten & Nicotra, 2011; Voskoboynik & Weissman, 2015). Botryllus schlosseri colonies propagate asexually by cyclical waves of blastogenesis (i. e. palleal budding) and during colony growth processes such as apoptosis and stem cell migration occur synchronously and massively (Rinkevich et al., 2013; Rosner et al., 2013; Franchi et al., 2016; Manni et al., 2019). In addition to its role as a model organism, Botryllus schlosseri is a cosmopolitan fouling species and one of the most widespread marine invaders (Berrill, 1950; Carlton, 2005; Ben-Shlomo et al., 2010; Bock et al., 2012; Lord, 2017), and is potentially able to damage the industry of shellfish aquaculture (Carman et al., 2010) and other marine industrial activities (Carver et al., 2006). In the context of the above-mentioned importance of this species, its taxonomy needs to be unequivocally and fully resolved, and the identification of the relative specimens has to occur beyond doubt based on clear diagnostic characters. However, in the last decade molecular data has shown that Botryllus schlosseri is a “ species complex ” consisting of five genetically highly divergent clades, named from A to E, each corresponding to a cryptic species (Lopez-Legentil et al., 2006; Bock et al., 2012). Clade A is globally distributed and clade E is present only in European waters, whereas the other clades are geographically restricted to a few localities (Stach & Turbeville, 2002; Lopez-Legentil et al., 2006; Perez-Portela et al., 2009; Bock et al., 2012; Yund et al., 2015; Nydam et al., 2017; Reem et al., 2017). The five clades have been identified for the first time by Lopez-Legentil et al. (2006) based on a short fragment of the mitochondrial (mt) gene cox 1, but they were not described as cryptic species but just as clades characterized by high nucleotide diversity, low haplotype diversity and restricted gene flow. Using both cox 1 and nuclear microsatellites, Bock et al. (2012) have put forward the hypothesis that these clades correspond to cryptic species. Indeed, Bock et al. observed a high interclade (10.8 – 16.5 %) and a comparatively low intraclade (0.8 – 3.8 %) cox 1 divergence and, at microsatellite level, a complete lack of contemporary gene flow between clades, suggestive of reproductive isolation (Bock et al., 2012). Later on, studies on the entire mitochondrial genome (mtDNA) identified three surprisingly distant taxa / subclades within clade A, that have been explained as further cryptic species or as the outcome of ongoing speciation events (Griggio et al., 2014). A Pacific specimen “ sc 6 ab ” from clade A has been the target of a genome sequencing project, thus leading to the publication of both the nuclear and mitochondrial genomes (Voskoboynik et al., 2013). Moreover, a Mediterranean clade A specimen having the same cox 1 haplotype of the mtDNA of the “ VE ” specimen (Griggio et al., 2014) has been designated as the neotype for Botryllus schlosseri when it was redescribed by Brunetti et al. (2017). The sc 6 ab and the VE specimens belong to two of the three divergent taxa / subclades identified by the mtDNA within clade A (Griggio et al., 2014). The exact phylogenetic relationships between the five clades remain elusive, because the few published cox 1 trees are not fully resolved, due to the presence of unreliably supported nodes (Lopez-Legentil et al., 2006; Yund et al., 2015) and / or were reconstructed with unusual and potentially inappropriate methodologies for the cox 1 nucleotide sequences (i. e. assuming a strict molecular clock (Yund et al., 2015) or analysing the translated sequences (Lopez-Legentil et al., 2006 )). However, without considering node reliability / methodology anomalies, all trees agree in recognizing clade E as sister to all other clades (Lopez-Legentil et al., 2006; Bock et al., 2012; Yund et al., 2015). Population genetic studies have also started investigating the geographic distribution, origin and dispersal history of the two widespread cryptic species, i. e. clade A and clade E (Bock et al., 2012; Yund et al., 2015; Nydam et al., 2017; Reem et al., 2017). Although the picture is far from complete, these studies have revealed that clades A and E are both successful invaders and have shown traces of recent range expansion of clade E (Nydam et al., 2017). Moreover, hypotheses on the geographic origin and dispersal history of the species complex / clades have also been presented (Berrill, 1950; Carlton, 2005; Nydam et al., 2017; Reem et al., 2017). Due to the relevance of this species complex, the need for a resolution on the taxonomic crypsis of this organism is growing. In particular, the Botryllus schlosseri taxonomic re-evaluation should lead not only to a more accurate discrimination of this species from other botryllids, but also to the identification of clear morphological and molecular diagnostic characters able to distinguish the various cryptic species. An accurate taxonomic description will help the studies on this model ascidian, with a better interpretation of possible differences in the outcomes of experiments carried out on different clades, as well as the finding of possible differences in the invasive potential of the various cryptic species, with consequences in the setting of the most appropriate monitoring and management strategies. For this purpose, and considering the lack of a type specimen and the previous unclear morphological descriptions, in 2017 we designated a neotype of Botryllus schlosseri, as reference point for future studies (Brunetti et al., 2017). Following an integrative taxonomy approach, this neotype was described both at the morphological and molecular level, allowing to associate a precise morphology to clade A (Brunetti et al., 2017) and in particular to a cox 1 haplotype identical to that of the previously published mtDNA of the VE specimen (European Nucleotide Archive [ENA] Accession Number: FM 177702) (Griggio et al., 2014). We believe that an integrative taxonomy approach should be applied to all genetically identified clades of the Botryllus schlosseri species complex with the aim of elucidating their taxonomic status. Here, in order to clarify the taxonomic status of clade E and its relationship to clade A, we analysed clade E specimens according to an integrative taxonomy approach that accounts for morphological and molecular characters. Then, we compared the clade E features to those of the Botryllus schlosseri neotype (Brunetti et al., 2017). In particular, we analysed the colony and zooid morphology, sequenced the entire mtDNA of a clade E specimen and compared several genome-level mitochondrial features (i. e. sequence divergence, gene order, and non-coding regions). The differences observed between clade E and clade A / Botryllus schlosseri sensu Brunetti (2017) at both the mitogenomic and morphological level show that clade E sensu Bock et al. (2012) can be recognized as a distinct species, which is described below.	en	Brunetti, Riccardo, Griggio, Francesca, Mastrototaro, Francesco, Gasparini, Fabio, Gissi, Carmela (2020): Toward a resolution of the cosmopolitan Botryllus schlosseri species complex (Ascidiacea, Styelidae): mitogenomics and morphology of clade E (Botryllus gaiae). Zoological Journal of the Linnean Society 190: 1175-1192
5947A847E948FFE83CF3C7F6F902FF73.taxon	description	E t y m o l o g y: Fr o m G a i a, t h e n a m e o f t h e f i r s t granddaughter of the author of this species, R. Brunetti. Material examined: Syntypes: MSNVE 25086, BT 2, colony fragment in 10 % formalin; MSNVE 2508, BT 4, colony fragment in 10 % formalin; MSNVE 25088, BT 2 A, another fragment of the BT 2 colony preserved in 99 % ethanol; MSNVE 25089, BT 4 A, another fragment of the BT 4 colony preserved in 99 % ethanol. All collected by F. Mastrotorato during clam sampling. Type locality: Barletta (Italy), 41 ° 22 ’ 55.646 ” N, 16 ° 10 ’ 06.586 ” E, 2.5 – 4.0 m depth on Cymodocea nodosa (Ucria) Asch. Seagrass meadow. Description: In both colonies from Barletta, labelled BT 2 and BT 4, the test surface is hard but, once it is torn, zooids are easily pulled out. Sample BT 4 consists of two pieces of colony, each of them with 2 – 3 systems at the beginning of the generation change (developmental stage 9 – 11 / 8 – 9 / 6 – 7 / 1) (Berrill, 1941). Therefore, the older generation is at the beginning of regression (stage 9 – 11) and the new generation presents syphons still closed and all internal organs completely developed (stage 8 – 9). The regressing zooids are clearly arranged in Botryllus schlosseri - type systems. Sample BT 2 contains a piece of colony of about 1 cm 2 of surface with two systems consisting of 15 zooids. As shown in Figure 5 A-B, in life these two colonies are green (BT 4) and green-yellow (BT 2), whereas in preservative there are whitish. The following description is chiefly based on the observation of the first order buds of sample BT 2, which have virtually finalized their development (only the syphons have to be completed) (Fig. 5 C – E). The adult regressing zooids are about 1.5 mm high and the same height is recorded in first order buds. Therefore, it is justifiable to deduce a height up to 2 mm for the filtering zooid, because an increase in size is observed when the bud opens the syphons. The body wall presents fine muscle fibres running from the anterior-dorsal side to the ventral one. In the buds there are eight tentacles: four large and four intercalated short. The atrial syphons, still visible in some regressing zooids, is conic with a dorsal tongue. The branchial sac has nine rows of stigmata on the right and eight on the left side; the second row is dorsally incomplete (Fig. 5 D – E). The zooid has a cylindrical shape having about 20 stigmata in the first half row and about 15 in the seventh one. The branchial formula at the level of fourth row of stigmata is usually E 5.5.5.5 DL. Along the transversal branchial vessels, no muscle fibres are detected. A simple edged dorsal lamina rises from the dorsal vessel at level of fourth to fifth row of stigmata and reaches a height of three times the diameter of the dorsal vessel (Fig. 5 D – E) at the oesophageal opening which is at level of the last row of stigmata. The intestinal loop is horizontally arranged (Fig. 5 D). The stomach is almost totally below the branchial sac. It is slightly cylindrical, with nine folds slightly spirally arranged and all posteriorly closed. These folds have a smooth surface (Fig. 5 F, I – L), without the longitudinal groove observed in Botryllus schlosseri (Brunetti et al., 2017: fig. 1). They are longer on the mesial side (folds 5 – 8 in Fig. 5 L) and without swellings at their cardiac end. The pyloric caecum is long, about half of the stomach, and has a rounded dilated tip; it rises from about the posterior third part of the typhlosole, the anterior part of which is much larger than the posterior one (Fig. 5 F, I). The intestine has a peculiar shape, with a hairpin bend (indicated with an arrow head in Fig. 5 C – D) at level of the first intestinal loop. This aspect is not due to pressure of the gonads, as it is present also in zooids without developed gonads. The rectum is strongly stretched by a connection with the branchial sac through a trabecula at the level of the transversal vessel between the seventh and eighth rows of stigmata (Fig. 5 D – E, H). The anus opens at level of the seventh row of stigmata, that is one or two rows above the oesophageal opening. The anal opening has four lobes, each of them splits into three minor ones (Fig. 5 G). On the rectum surface, there are four grooves going from the anus to the pyloric gland. In sample BT 4, well-developed gonads are present only in buds of the first order, which are ready to substitute the parents. There are up to four eggs (diameter of about 140 μm) on each side, anterior and dorsal to the testis follicles, which partially embrace them as a cup (Fig. 5 E). Gonadal primordia are present also in buds of the second order. No larvae and embryos have been found. Remarks: Based on the structure of the cloaca, this species clearly belongs to the genus Botryllus, as defined by Brunetti (2009). The main morphological features distinguishing Botryllus gaiae from Botryllus schlosseri are reported in Table 4 and include several traits currently considered valid diagnostic features to discriminate species in the genus Botryllus (Brunetti, 2009; Brunetti & Mastrototaro, 2017). These traits, highlighted in bold in Table 4, concern peculiarities of the stomach and the intestine, such as the shape of the first and second intestinal curve, the shape and position of the anus opening, the number and surface appearance of the stomach folds and the shape of the typhlosole. All these morphological differences are highly significant and support the hypothesis that clade E is a new species, distinct from Botryllus schlosseri sensu Brunetti et al. (2017) and therefore from clade A.	en	Brunetti, Riccardo, Griggio, Francesca, Mastrototaro, Francesco, Gasparini, Fabio, Gissi, Carmela (2020): Toward a resolution of the cosmopolitan Botryllus schlosseri species complex (Ascidiacea, Styelidae): mitogenomics and morphology of clade E (Botryllus gaiae). Zoological Journal of the Linnean Society 190: 1175-1192
