Diversity of Bugulidae (Bryozoa, Cheilostomata) colonizing artificial substrates in the Madeira Archipelago (NE Atlantic Ocean)

Until very recently the Madeira Archipelago registered a total of eight Bugulidae species. In the present study we include descriptions of seven Bugulidae species, now with Scanning Electron Microscopy images, with four new records for the Archipelago: Bugulina fulva and Bugulina simplex for Madeira Island and Bugula neritina and Crisularia gracilis for the neighbouring Island of Porto Santo. Furthermore, we report the correction of the previous identification of Bugulina calathus minor earlier reported from Funchal harbour in 1998, now as Bugulina flabellata. This study is part of an ongoing monitoring program for detecting non-indigenous species in all marinas from the Madeira archipelago, between 2007 and 2015. Specimens were collected in previously deployed PVC settling plates, marina pontoons, and also on recreational hull vessels while performing dry dock inspections at a local shipyard. Our study reveals that the Madeira archipelago now registers a total of ten Bugulidae species, contributing therefore to the total bryozoan fauna of the Archipelago, now with more than 100 records. These numbers could increase, as Madeira is considered to be a “hotspot” for bryozoan diversity when compared to other nearby regions. Finally, hull fouling is considered as the most likely vector of introduction for the non-indigenous species of Bugulidae detected in Madeira.


Background
The introduction of non-indigenous species (NIS) remains today a global threat to biodiversity in coastal marine ecosystems [e.g. 1,2]. This phenomenon has significantly increased in the past years as maritime traffic represents one of the main vectors contributing to the spread and establishment of NIS worldwide, facilitating the transport of species in ship hulls and ballast water [3][4][5]. Floating docks and other artificial substrates are very common in bays and estuaries, particularly in shipping areas and marinas are highly relevant for detecting new NIS arrivals [6][7][8]. The occurrence of artificial pontoons around the world may favour the dispersal of NIS, by providing stepping stones that allow colonization into new areas [9,10]. In fact, marinas and harbours have become an ideal study system for rapid assessment surveys (RAS) of NIS [e.g. 7, [11][12][13].
The Madeira archipelago is located Southwest of continental Europe and 700 km off the Moroccan coast, with a strategic geographic position in the Atlantic Ocean. Historically, the archipelago has provided an important passage route for many ships between Europe and the Caribbean, American and African continents. Today most of the maritime traffic comes from tourist cruise ships and sailing yachts for re-fuel, and rest stops from different parts of the world [14]. The diversity and distribution of fouling NIS in Madeira Island, North-eastern Atlantic Ocean, has received some attention in recent years due to on-going monitoring surveys, particularly in marinas on the southern coast of Madeira Island. As a result, several new species have been detected and inventoried [15][16][17][18], including new undescribed bryozoan species [19].
The phylum Bryozoa, (Ectoprocta) often known as moss animals, is a very diverse group with approximately

Open Access
Helgoland Marine Research *Correspondence: patramalhosa@gmail.com 1 MARE -Marine and Environmental Sciences Centre, Estação de Biologia Marinha do Funchal, Cais do Carvão, 9000-107 Funchal, Madeira, Portugal Full list of author information is available at the end of the article 5900 living species [20,21] and 15,000 fossil records [22]. Bryozoans are mostly sessile, colonial, filter-feeder organisms that can be found living on a variety of different natural and artificial substrates such as rocks, floating pontoons, buoys or ropes [19,23].
In the mid nineteenth and early twentieth centuries the pioneer work of Busk [24][25][26][27][28], Hincks [32], Johnson [31], Waters [30], and Norman [29] in surveying the bryozoan fauna of the shallow waters of Madeira resulted in a list of 139 species. Since then, a few species were added to this list, with most of the new records detected in the twentieth century [33][34][35][36]. In recent years, new records of bryozoans have been reported for Madeira Island [15,17,19,37]. Nevertheless, Berning [38] indicates that the known bryozoan species list for the Madeira archipelago is likely far from reflecting the true species richness for this group when compared to other coastlines. The author further considers the 250 km coastline of the Madeira archipelago a ´hotspot´ of bryozoan diversity, when compared to the 260 species known from the 1600 km of the Galician coastline [38].
The present study aims to review a complete list of the Bugulidae species colonizing artificial substrates in the Madeira archipelago. In this paper, we include descriptions of seven Bugulidae species with Scanning Electron Microscopy (SEM) illustrations, including four new records: Bugulina fulva [47] and Bugulina simplex [53] for the Madeira Island and Bugula neritina and Crisularia gracilis for the neighbouring Porto Santo Island. Furthermore, we report here the correction of the previous identification of Bugulina calathus minor earlier reported from Funchal harbour in 1998, now as Bugulina flabellata [44] for the Madeira Island.

Methods
In 2006 we initiated a NIS monitoring program in a local marina located along the South-eastern coast of Madeira Island [17,54]. We later expanded this study by adding a second marina in 2010, and two additional marinas in 2013 [16,18]. Currently, our NIS monitoring program covers all four main marinas of the archipelago: Calheta (32°43′N, 17°10′W), Funchal (32°38′N, 16°54′W), Quinta do Lorde, Caniçal (32°44′N, 16°42′W), and Porto Santo Island (33°03′N, 16°18′W) ( Fig. 1a-d) [16]. At every 3 months, plates were retrieved from the field and examined to determine species richness. During each sampling event, plates were carefully examined using a stereo microscope (Leica Wild-M3 Heerbrugg), and specimens from settling panels were photographed using a Sony DSC-W55 camera. Samples were collected from different marina sites (Fig. 1a- Additionally to settling plate deployment, we also initiated in 2013 dry dock inspections (DDI) on yacht hulls at a local shipyard (REPMaritima, www.repmaritima.com) for monitoring NIS arrivals via hull fouling. The ship's hull was photographed after being raised from water, at three different areas (bow, beam and stern) with a quadrate (20 × 20 cm), using a Canon A620 camera. Species inspections were performed by personal observations, and unknown species samples were taken for later species identification. Furthermore, we collected further information concerning vessel history from yacht owners through questionnaires. These questionnaires aimed at getting information regarding vessel identification (i.e. name, nationality, home port, hull material and length of the boat), last port of call, dry docking (i.e. how often does the boat goes into a dry dock and whether fouling was removed), and hull maintenance history (i.e. which antifouling paint was last used and when was the last antifouling renewal). In this context, we have surveyed a total of nine vessels from 2013 to 2015 where five were foreign with homeports in the North Sea and North Atlantic, while four were domestic.
Bugulidae specimens collected from harbours and DDI were later examined with scanning electron microscopic (SEM) photographs taken in a FEI Inspect S50 SEM using uncoated material at the University of Vienna, Austria. SEM was used with a back-scattered electrons detector in low vacuum mode. Part of the samples was exposed to an ultrasound bath, bleached by sodium hypochlorite and air dried for SEM examination. Measurements (in mm) were taken with the software ImageJ ® [56] on the SEM photographs and are indicated in the descriptions of each species. Average measurements are included in mm while minimum, maximum data, and the number of measurements attained are included between brackets.
Finally and to complement this study, we conducted a comprehensive literature survey on previous records of Bugulidae species present in the Madeira archipelago. All species detected in settling plates, DDI and found in literature surveys were assigned to one of three categories of biogeographically distribution: native, NIS or cryptogenic (unspecified native origin, and unknown whether native or NIS), in accordance to literature and several current databases [57][58][59][60].
However, Bugulina calathus minor (sample MMF31648), earlier reported from Funchal harbour in 1998 by Alves and Cocito [36], was carefully examined during this study and is now recognised as co-specific with our B. flabellata samples. Nevertheless, species found in our literature survey were not detected in our samples except for C. gracilis and B. flabellata. Consequently, B. flabellata was therefore detected in Funchal harbour in 1998 [36],  Table 1). Finally, as a result of our dry dock inspections, we have found records of four Bugulidae species: B. neritina, B. stolonifera, B. fulva and B. simplex. The most prevalent of these species was Bugula neritina, found in seven vessels (in four foreign vessels and in three domestic vessels). Bugulina stolonifera was present in one recreational vessel with Northern Europe origin. Finally, Bugulina fulva and Bugulina simplex were both detected in one local recreational vessel from Funchal. Next, we describe the seven Bugulidae species detected in artificial substrates in the Madeira archipelago from 2013 to 2015:

Diagnosis
Erect colony, branching, forming dense tuft until 10 cm high. Colony brown reddish, sometimes even violet or purple. Biseriate branches, with rectangular boat-shaped autozooids, with almost the entire frontal wall membranous. Autozooids without spines but with the distal margin well marked. Avicularia absent. Big and globular ovicells jointed by a peduncle to the maternal zooid and with oblique orientation.

Remarks
Bugula neritina is a well-known widespread species, it was first detected for Madeira Island by Norman [29] in 1909 in Funchal, near a coaling ship in natural substrate. It has not been reported elsewhere, up until 2012 where it was found growing on the artificial PVC settling plates  (Table 1).
In addition, we have found B. neritina present in seven of the nine recreational vessels examined for DDI, in four foreign vessels, and in three domestic vessels from Funchal.

Description
Colonies erect, forming dense tufts, light yellow in colour. Branches formed by series of three to eight autozooids in width. Autozooids elongated, rectangular, about 0.612 mm (0.499-0.763 mm, 19) long and about 0.148 mm (0.130-0.182 mm, 19) wide. Frontal membrane occupies almost the entire frontal area. Spines very variable in size along the colony. Three spines in the outer distal corner, the two of them more proximal are directed obliquely or frontally, and the one more distal is directed distally. Two spines in the inner distal corner, both with disto-frontal direction. Pedunculate avicularia positioned in the distal part of the margin of the frontal membrane, below the spines. Avicularia on the outer zooids big in size, about 0.187 mm (0.154-0.218 mm, 16) and smaller in the inner zooids 0.106 mm (0.094-0123 mm, 11). The beak of the avicularia abruptly, almost rectangular, hooked. Globular big ooecia located in the midline of the maternal zooid. Embryos light yellow in colour.

Remarks
Busk [24] recorded Bugula flabellata var. biseriata s. ditrupae from Madeira, and subsequently the same author [25] identified it as a different species, Bugula ditrupae and later Norman [29] confirmed this record. Both species are easily distinguishable, Bugula ditrupae is formed by biserial branches and in Madeira it was recorded always associated with species of the genus Ditrupa (polichaeta) [29].
Alves and Cocito [36] recorded Bugulina calathus minor from Madeira Island in 1998. Nevertheless, we examined the specimen studied by Alves and Cocito [36] (MMF31648) and it seem co-specific with our material. It presents the beak of the avicularia hooked almost rectangularly, as it happens in our specimens (can also see in Alves and Cocito Fig. 1a, note that it is not possible to see in the Fig. 1b because the position of the avicularia), been the beak of the avicularia smoothly down-curve in B. calathus [e.g. 47,62]. Beside, in the avicularia of the specimens from Alves and Cocito [36] have the edge from the peduncle to the mandibular proximal border shorter and more curve than in the avicularia of Bugula calathus [e.g. 47,51,62,63].
No other records of this species have been detected for Madeira Island until this work; here, we found B. flabellata for the first time for the marina of Calheta in 2013 and 2014, and for the marina of Funchal in 2014 and 2015.

Remarks
To date, Bugulina fulva has never been recorded for both Madeira and Porto Santo Islands. This study represents the first record of this species for Madeira Island with records for the Funchal marina in 2013 and 2014. This species was also found in one local recreational vessel from the Funchal marina and still remains undetected elsewhere in Madeira.

Description
Colony erect, funnel shape, not-spiralling, up to 3 cm in height, translucent white to yellow in colour. Branches formed by series of two to six autozooids, generally narrow at their origin and broader distally. Autozooids

Remarks
Bugulina stolonifera was first reported for Madeira in the marina of Quinta do Lorde in 2010 by Canning-Clode, Fofonoff [17]. In 2014, this species was still present in the same location. Also in 2013, we detected its presence during a DDI in a French origin vessel. In 2014, we recorded the presence of Bugulina stolonifera for the marinas of Funchal and Quinta do Lorde.

Description
Colony erect, with spiral growing and up to 5 cm high, yellow pale in colour, with dicotomical branches and  (Harmer, 1923). Elongated autozooids, 0.569 mm (0.476-0.662 mm, 9) long and 0.139 mm (0.105-0.162 mm, 9) wide; the frontal membranous occupies between two-thirds and one half of the frontal area, with a U shaped proximal edge. Outer distal corner has generally a short and slender spine; inner side with a longer spine; and one-third generally longer and slender spine, about 0.230 mm (0.177-0.280 mm, 6), situated in the middle of the distal margin, slightly moved to the outer side. Pedunculate avicularia long, with a well marked indentation between the frontal body surface and the rostrum; and presents the beak steep down curved. The avicularias are few abundant, disposed in the outer margin situated in the medium of the distance of the membranous area. Ooecia globular with only a small distal area calcified.

Remarks
Crisularia gracilis was first described from Madeira by Busk [24] in 1858, and it is here described for the first time for Porto Santo Island. This specimen was collected in 2015 at one of the pontoons at Porto Santo marina at 0.5 m depth and not from the PVC panels deployed at the marina. The identity of Crisularia gracilis is confused [see 65] but our specimen seems to coincide with the original and other subsequent descriptions found in the literature [24,62,65]. Previous records of this species seem questionable and in some cases have been assigned to other species (see Bugula turrita in Ryland [47]).
Biogeography status: due to the problems in the identification and identity of this species [see 65] and as it was originally described from Madeira [24], we considered it here as a cryptogenic species, however additional work will be necessary to clarify the status of this species.

Remarks
This species was recorded for the first time in Madeiran waters by Norman [29] in 1909, in Porto Santo Island. More than a century later, Canning-Clode, Fofonoff [17] have detected Virididentula dentata (as Bugula dentata) for Madeira Island in Quinta do Lorde marina in 2007, 2010 and 2012. Since then, it has been detected in both Quinta do Lorde and Porto Santo marinas in 2013, 2014 and 2015.

Discussion
Bryozoans represent a group of colonial invertebrates in benthic ecosystems and can be commonly found in fouling assemblages [66,67]. The transportation of fouling communities in ship hulls is considered a significant and ancient vector for the introduction of non-indigenous species at a global scale [68,69]. Like other fouling communities representatives (e.g. barnacles, mussels) bryozoans seem to present life forms that make survival in ship hulls possible [70].
Bugula neritina is a widely known cosmopolitan bryozoan species since the eighteenth century [71], distributed in warm-temperature and subtropical coastal waters, known as an important component of the fouling communities colonizing artificial substrates (see historical review in Ryland, Bishop [50]). This species is able to attach to ship hulls [41] showing tolerance to heavy metals, such as copper and zinc present in several antifouling paints [55,72]. Bugula neritina is currently considered a species complex forming at least three different lineages (Type S, D and N) that also differ in their distributional patterns [49,71,73,74]. Type S is the most widely distributed, appearing in tropical, subtropical and temperate waters around the world, including the North-eastern Atlantic [49,50,71]. Type D has a more restricted distribution, probably endemic to California [71,73] while Type N was initially found only in the North-western Atlantic [74] but latter Fehlauer-Ale, Mackie [71] found this type in Townsville (Coral Sea) and Central California, what could indicate a recent expansion. In this context, we presume the specimens of B. neritina found in Madeiran waters belong to type S, but we emphasize the need for future molecular studies to confirm this assumption.
Linnaeus [45] described this species for the Mediterranean and the American seas, but the true origin of B. neritina remains unknown. According to the dispersal capacity of a sessile species with a lecitotrophic larva [75,76] it is not expected the actual distribution of Type S for natural dispersion. Several haplotypes were discovered with a long distribution of the Type S, with supported correlation with sea surface temperature (for more details see Fehlauer-Ale, Mackie [71]). The study of these different haplotypes allow some latitude in speculating about the possibility of two different origins of Type S for B. neritina, the North-eastern Pacific (California) or Southwestern Atlantic (Brazil) [71].
In Madeira, Bugula neritina was first recorded by Norman [29] in Funchal, near the coaling ship inside the port area which today no longer exists, since it was destroyed in the late 1960s. Surprisingly, until recently it has not been reported elsewhere for Madeira, perhaps due to a low sampling effort over the years. However, since we initiated our surveys for NIS diversity in local marinas, B. neritina was detected on artificial PVC plates in the Calheta marina in 2012. Moreover, from 2013 to 2015 B. neritina was always found present in all four marinas of the Madeira archipelago, including a first record for the Porto Santo marina in 2013. It is not possible to determine since when B. neritina has been present in Porto Santo Island, but we suspect this species has been likely introduced to both Madeira and Porto Santo via hull fouling. Bugula neritina is considered here as NIS [50,[58][59][60]. Our data from dry dock inspections indicate a significant prevalence in the hull of recreational vessels, as it was found present in four foreign vessels and three local recreational vessels from Funchal.
Bugulina flabellata is considered native and widely distributed in the Western Europe [50], and was also recorded in the Mediterranean [77] and Adriatic seas [78]. Bugulina flabellata was recently documented as introduced in Australia and New Zealand, presumably having travelled on ships [50,79]. In this study, we re-examined the material collected by Alves and Cocito [36] and have corrected its identification to B. flabellata instead of Bugulina calathus minor, since it is co-specific with our sampled material. This is still the first record for Madeira Island which was collected from Funchal harbour in 1998 by Alves and Cocito [36]. Since then, we detected B. flabellata in 2013 and 2014 for the marina of Calheta and yet again, for the marina of Funchal in 2014 and 2015. Given its broad distribution in European waters, we considered it as being native in Madeira [47,58].
Bugulina fulva was first described from the England Coast [47] occurring in natural (holdfast of Laminaria and boulders) and artificial substrates, appearing during summer in England and Wales [63,80]. After being described, Bugulina fulva was recorded from different parts of the world. Maturo [81] found this species at the South coast of the USA of America always on natural substrates. Nevertheless, this species is characterized to appear as a fouling component on artificial substrates: was found from the tropical west coast of Africa in the Luanda harbour (Angola) by Cook [82]; in the western Mediterranean by Zabala [83]; and in the Adriatic [see 78]. Also, [84,85] found this species in Belgian waters colonizing plastic debris. Ronowicz, Kukliński [80] found this species as a component fouling assemblages during summer settlement experiments. Hayward and Ryland [63] also recorded this species in the West Atlantic coast from Maine to Brazil. However, no previous records were found for this species in Madeira Island. According to the general distribution of this species, we also believe it could reflect a recent introduction for the Madeira region and is assigned here as cryptogenic [57][58][59].
Bugulina simplex is well known from settlement panels used for fouling studies in Britain, but was never found on natural substrates by dredging [47]. In the USA (Woods Hole, Massachusetts), it was found mainly on pier piles and other submerged structures [86], and only rarely colonizing macroalgae. Nevertheless, [87] found B. simplex to be abundant on communities of Zostera marina at 1-6 m depth and in the aquaculture areas of oysters, present both on oysters and on the cages in the Mediterranean. Bugulina simplex has been characterized as a fouling species [47] and recognized as NIS worldwide [50]. Bugulina simplex appears to have a Mediterranean origin [50] and its presence in Madeiran waters is probably a result of a recent introduction and assigned here as NIS [50,58,60].
Bugulina stolonifera was first described by Ryland in 1960 from England, mainly found colonizing pier pilings and hull vessels [47]. Cohen and Carlton [88] considered this bryozoan as probably being native to the Northwest Atlantic. In Ryland, Bishop [50] Bugulina stolonifera is recognized as invasive worldwide. Bugulina stolonifera was first described and assigned for Madeiran waters in 2010 as NIS [17,50,58].
Finally, a new genus was recently introduced into the bryozoan literature to accommodate the species Virididentula dentata [42], based on molecular analysis and its peculiar morphological characters. This species was described by Lamouroux [46] from Australasia, and latter different records were confirmed in other regions such as Australia and New Zealand [e.g. 89,90], South Africa to Cape Verde [e.g. [91][92][93], Hawaii [48], and Brazil [94]. This broad distribution; the intraspecific morphological variation, mainly related with the zooidal position morphology of the avicularia (no giant avicularias were observed in our specimens); and the orientation of the ooecia [42,48,90] could be indicative of a species complex [42,48]. Some authors consider the Indo-Pacific as the origin of this species [17,95], but other authors described it as a cryptogenic species [94,96]. For the Madeira region V. dentata is assigned as NIS [17].
Crisularia gracilis was first described from Madeira by Busk [24] in 1858, and is here described for the first time for Porto Santo Island in 2015. Given its original description locality [24] and the lack of more studies we considered Crisularia gracilis as being a cryptogenic species.
Prior to 2012, eight Bugulidae species were recorded for the Madeira Archipelago. During our surveys conducted in the four marinas (Calheta, Funchal, Quinta do Lorde and Porto Santo), we found no evidences for the occurrence of three of these species: Bugulina avicularia, Bugulina ditrupae, and C. plumosa. However, we cannot exclude the possibility of their occurrence in other artificial or even natural substrates, inside or outside marinas and harbours in Madeiran waters. Furthermore, we cannot exclude the possibility of mis-assigned species from earlier records, as it was seen here for Bugulina flabellata as Bugulina calathus minor. Unfortunately, it was not possible to confirm the identity of previously collected specimens of Bugulina avicularia, Bugulina ditrupae, and C. plumosa (and found in our literature survey) as these are not easily accessible, and therefore we suggest additional studies to confirm the identity of these species. Nevertheless, we attributed a native status to Bugulina ditrupae, originally described from Madeira by Busk [24], while Bugulina avicularia and C. plumosa were considered cryptogenic [29].
The present paper represents therefore a key contribution to the current understanding of bryozoan species of the Madeira Islands, regarding Bugulidae. The Madeira archipelago has now registered a total of ten Bugulidae species, with three species present both in Madeira and Porto Santo (B. neritina, C. gracilis and V. dentata), six exclusively found in the Madeira Island (B. avicularia, B. ditrupae, B. flabellata, B. fulva, B. simplex, and B. stolonifera), and one exclusively found in the Porto Santo Island (C. plumosa). Our preliminary findings on dry dock inspections on local and foreign recreational vessels, support the hypothesis from other studies, that emphasize hull fouling as a key introduction vector that contributes to the arrival and spread of NIS to neighbouring regions. Thus, we strongly believe that all non-indigenous species of Bugulidae reported here for the Madeira archipelago were introduced via hull fouling.