INTRODUCTION
The family Thorectidae Bergquist, Reference Bergquist1978 (Demospongiae: Dictyoceratida) is distributed throughout the world from tropical to temperate seas, except for polar regions, and from the intertidal zone to around 100 m depth (Cook & Bergquist, Reference Cook and Bergquist1996, Reference Cook, Bergquist, Hooper and Van Soest2002). It includes 23 valid genera and approximately 130 species whose shape varies from encrusting pads to massive, erect, lobate, caliculate or digitate growth forms. The surface is usually armoured with a sandy cortex or sculptured with ridges, pits or conules. The skeleton is a regular reticulum with rectangular meshes formed by primary, secondary and sometimes tertiary spongin fibres. Fibres are laminated in cross-section, with clear zones of disjunction between successive layers and a diffuse pith that may be obscured by an axial core of foreign material. Some fibres may be very stout or form complex fascicles (Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002).
The genus Thorecta Lendenfeld, Reference Lendenfeld1889 is the type genus of Thorectidae. It is currently considered to comprise 20 valid species (Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002), although the exact number of valid species is controversial and varies from 8–24 according to the classifications adopted by different authors (e.g. Lendenfeld, Reference Lendenfeld1889; de Laubenfels, Reference De Laubenfels1948; Wiedenmayer, Reference Wiedenmayer1977, Reference Wiedenmayer1989; Bergquist, Reference Bergquist1980; Cook & Bergquist, Reference Cook and Bergquist1996, Reference Cook, Bergquist, Hooper and Van Soest2002). Species of Thorecta show stalked, globular, fan-shaped or cylindrical upright growth forms. The surface is always armoured with a sandy cortex; it may be smooth or regular, but not with typical conules; some species however may have microconulose areas on the surface. The reticulate skeleton has regular rectangular meshes, formed by primary fibres cored with debris and clear secondary fibres (Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002; Cook, Reference Cook, Custódio, Lôbo-Hajdu, Hajdu and Muricy2007). Most species referred to the genus are distributed in the Indo-Pacific, especially in Australia (19 species), but also in Micronesia, Madagascar, Mauritius and New Zealand (one species each).
Six species assigned to Thorecta were previously recorded from the Atlantic Ocean, viz., Thorecta dendroides (Lendenfeld, Reference Lendenfeld1889), Thorecta galeiformis Lendenfeld, Reference Lendenfeld1889, Thorecta gracillimus Lendenfeld, Reference Lendenfeld1888, Thorecta laxus Lendenfeld, Reference Lendenfeld1889, Thorecta horridus (Hyatt, Reference Hyatt1877), all from the Bahamas, and Thorecta squalidus Lendenfeld, Reference Lendenfeld1888 from South Africa. The validity of all these species, however, is highly questionable: T. dendroides and T. laxus are considered unrecognizable (de Laubenfels, Reference De Laubenfels1948; Bergquist, Reference Bergquist1980); T. galeiformis is considered a junior synonym of T. calyx, and its records from the eastern Pacific and Caribbean are doubtful (de Laubenfels, Reference De Laubenfels1948; Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994); T. gracillimus is a junior synonym of Taonura colus (cf. Bergquist, Reference Bergquist1980; Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994); T. horridus is probably the poecilosclerid Mycale (Arenochalina) laxissima (Duchassaing & Michelotti, Reference Duchassaing de Fonbressin and Michelotti1864) (see van Soest, Reference Soest, Hummelinck and Van der Steen1978; Zea, Reference Zea1987; Wiedenmayer, Reference Wiedenmayer1989); and T. squalidus was transferred to Phoriospongia (Poecilosclerida; Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994). In Brazil, Muricy et al. (Reference Muricy, Santos, Batista, Lopes, Pagnoncelli, Monteiro, Oliveira, Carvalho, Melão, Moreira, Klautau, Rodriguez, Costa, Silvano, Schwientek, Ribeiro, Pinheiro, Hajdu, Lavrado and Ignacio2006) recorded without description a Thorecta sp. and other unidentified Thorectidae from Bahia State, and Muricy et al. (Reference Muricy, Esteves, Moraes, Santos, Silva, Klautau and Lanna2008) described a Thorecta sp. from Rio Grande do Norte State, without naming it or designating type specimens. In summary, no valid species of Thorecta have been described from the Atlantic Ocean so far.
In the present paper we describe in detail the specimens of Thorecta sp. from Rio Grande do Norte State (Muricy et al., Reference Muricy, Esteves, Moraes, Santos, Silva, Klautau and Lanna2008) as a new species, T. atlantica sp. nov. The new species is the only currently valid species of the genus described in the Atlantic Ocean so far. We also discuss the taxonomy and species composition of the genus Thorecta.
MATERIALS AND METHODS
Potiguar Basin is located on the north continental shelf of Rio Grande do Norte and Ceará States (north-eastern Brazil). The sampling area ranged between 04°30′00″–05°10′00″S and 36°10′00″–36°50′00″W, off the cities of Guamaré, Galinhos, Macau and Areia Branca, in Rio Grande do Norte State (Figure 1). Potiguar Basin is an area of oil and gas exploitation in which the research centre (CENPES) of the Brazilian oil company PETROBRAS is carrying on a programme of environmental characterization and monitoring, including a description of the sponge fauna (Muricy et al., Reference Muricy, Esteves, Moraes, Santos, Silva, Klautau and Lanna2008).
Fig. 1. Location of the studied area, Potiguar Basin, and of the collection sites of Thorecta atlantica sp. nov. (black circles in C).
The sponges studied here were collected by trawling on board of the RV ‘Astro Garoupa’ in three campaigns (BPOT 2–4) by the Project of Environmental Characterization and Monitoring of Potiguar Basin (Muricy et al., Reference Muricy, Esteves, Moraes, Santos, Silva, Klautau and Lanna2008). Dredging and trawling were carried out in 136 stations within the study area, but the new species of Thorecta was found in only two of them (Figure 1C). Trawling was performed with a net 18 m long with a mouth of 9 m, mesh diameter of 30 mm in the body and 24 mm in the sac. Samples were fixed and preserved in 70% ethanol. The specimens were deposited in the Porifera collections of the Departamento de Zoologia da Universidade Federal de Pernambuco (UFPEPOR) and Departamento de Invertebrados do Museu Nacional, Universidade Federal do Rio de Janeiro (MNRJ). Thick section preparations followed standard procedures (Cook, Reference Cook, Custódio, Lôbo-Hajdu, Hajdu and Muricy2007). The measurements of fibre diameter are presented as minimum–mean–maximum; N = 20 per individual. Photomicrographs were taken with a Sony DSC-W50 digital camera coupled with a Nikon Eclipse E-200 light microscope.
RESULTS
TYPE SPECIES
Thorecta exemplum Lendenfeld, Reference Lendenfeld1888 var. tertia Lendenfeld, Reference Lendenfeld1888 (by subsequent designation; de Laubenfels, Reference De Laubenfels1948); junior synonym of Thorecta latus (Carter, Reference Carter1885) (cf. Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002).
DIAGNOSIS (FROM COOK & BERGQUIST, Reference Cook, Bergquist, Hooper and Van Soest2002)
Stalked, globular, fan-shaped or cylindrical upright growth forms, in which the primary fibres are cored with a regular axial column of debris and the secondary fibres are clear. The skeleton forms a regular, almost perfectly rectangular mesh in which the spaces between the fibres may be quite large (2 mm is common). The surface is always armoured, not conulose. These sponges are firm and compressible.
Thorecta atlantica sp. nov.
TYPE MATERIAL
Six specimens, all from Potiguar Basin, Rio Grande do Norte State, Brazil, coll. RV ‘Astro Garoupa’—holotype: UFPEPOR 360, Campaign BPOT 04, Station A22, off Guamaré, (04°47′50.8″S–36°18′16.8″W), 62–63 m depth, 24 May 2004 (Figure 2A). Paratypes: UFPEPOR 121, Campaign BPOT 02, Station A4, off Porto do Mangue (04°37′31.7″S–36°46′00.7″W), 70–101 m depth, 14 May 2003; UFPEPOR 247, Campaign BPOT 03, Station A4, off Porto do Mangue (04°37′09.4″S–36°44′57.1″W), 61–160 m depth, 14 November 2003; UFPEPOR 442 (Figure 2B–D), 667, 668, Campaign BPOT 04, May 2004, Potiguar Basin; precise locality, depth and date of collection unknown; schizotype: MNRJ 13170 (fragment of UFPEPOR 442).
Fig. 2. Thorecta atlantica sp. nov. (A) Holotype (UFPEPOR 360), oblique view; (B) paratype (UFPEPOR 442), upper view; (C) another paratype (UFPEPOR 247), longitudinal section; (D) close-up of the surface showing the irregular openings. Scale Bars: A = 5 cm; B, C, D = 1 cm.
DIAGNOSIS
Thorecta with globular or clavate (stalked) shape and 1–2 large apical oscules with deep atria.
SYNONYM
Thorecta sp., Muricy et al., Reference Muricy, Esteves, Moraes, Santos, Silva, Klautau and Lanna2008: 115.
DESCRIPTION
External colour greyish-brown to dark brownish-grey, sometimes with whitish or yellowish tinges in some parts of the surface, and with the lowest portion often in a lighter shade of brown; the interior is always brown. Shape massive, clavate, globular, ovoid or subspherical, with 1–2 large apical oscula and often fixed in the substrate by 1–2 thick peduncles (Figure 2A–C). Two specimens were fragments of sponges cut in half longitudinally, and three are damaged and apparently have lost their peduncles during collection by dredging. The largest specimen (the holotype) measures 27 cm high by 22–24 cm in diameter and has irregular lobes 23–45 mm high by 16–41 mm wide in its lower portion; it has two large oscula on top, 4.0–6.5 cm in diameter, each one with a large atrium 9.5–12 cm deep which becomes gradually thinner towards the base; the peduncle is 8.5 cm high by 7–8 cm in diameter, and it is subdivided in its base in four smaller extensions 3–4 cm high by 2–3 cm in diameter (Figure 2A). The smallest specimen is irregularly ovoid, 8 cm high by 6 cm in diameter, with a single sub-apical oscule, 1.5 cm in diameter, with a shallow atrium 3.7 cm deep; the sponge is fixed by two irregular peduncles, 20–21 mm high and 12.4–26.3 mm in diameter. The other specimens were similar or intermediate between these two in most characters except in that their apical oscules have an elevated rim, 3–13 mm high, and some have lost their peduncles.
Surface smooth or rugose, formed by a detachable, armoured cortex. In parts where the dermis has been lost, the surface is pierced by the extremities of the primary tracts, which form irregular crests or pointed elevations 1–9 mm wide by 2–4 mm high. These crests are sometimes aligned, forming radially arranged ridges. The surface has large patches of numerous, closely-spaced irregular openings, 0.7–4.8 mm in diameter (Figure 2D). In some parts these openings fuse together forming elongate depressions, 23–25 mm long by 2–4 mm wide. Consistency is hard but elastic in the choanosome and firm, resistant in the cortex. The peduncle is harder than the main body and incorporates large and abundant inorganic debris (fragments of shells, calcareous algae, etc.).
Dermal armour cortex is regular, granular, 150–300 µm thick (Figure 3A). Skeleton is reticulate, with strongly lamellate fibres, clearly distinguishable in primary and secondary ones (Figure 3B). In the ectosome the reticulation is regular, with ascending primary fibres 80–179–300 µm thick cored by relatively small amounts of detritus (Figure 3B, E). Secondary connective fibres, 24–101–228 µm thick, are free of detritus (Figure 3B, F). Meshes are almost rectangular, 60–2400 µm wide. Choanosome cavernous, with large canals 1.5–7.5 mm in diameter arranged more or less radially. Choanosomal reticulation more irregular, with rounded or polygonal meshes; choanosomal primary fibres often free of detritus, hardly distinguishable from secondary fibres (Figure 3C). Peduncle with an irregular reticulation in cross-section and longitudinal fibres tightly packed in transverse sections (Figure 3D); coarse sediment is very abundant, making the consistency there more rigid than in the main body of the sponge. Intraspecific variability in morphometric characters is relatively low (Table 1).
Fig. 3. Thorecta atlantica sp. nov. (A) transverse section of the ectosome (S, surface); (B, C) transverse sections of the choanosome (surface is in the right); (D) skeleton of the peduncle, heavily loaded with detritus; (E) primary fibre; (F) secondary fibre. Scale Bars: A = 800 µm; B, C, D = 300 µm; E, F = 50 µm.
Table 1. Intraspecific variation of morphometric characters of Thorecta atlantica sp. nov. (in µm). F1, primary fibres; F2, secondary fibres.
ETYMOLOGY
The adjective atlantica derives from the distribution of this species, which is the first valid species of Thorecta described from the Atlantic Ocean.
DISTRIBUTION
Only known from Potiguar Basin (Rio Grande do Norte State, north-eastern Brazil).
ECOLOGY
Hydrozoans and crustaceans were found associated with most specimens. This species occurs in relatively deep water, from 61–160 m depth.
DISCUSSION
The new species clearly belongs to Thorecta as demonstrated by its stalked globular shape, smooth surface, regular dermal armour, cored primary and uncored secondary fibres, and absence of mucus. It is easily characterized within the genus by its shape (massive globular or clavate) with one or, most often, two large apical oscules with deep atrial cavities. Among the species that we suggest that should remain in Thorecta, T. carteri, T. farlovi, and T. polygonum differ from the new species by the flabellate shape. Thorecta vasiformis is cup-shaped; T. marginalis, T. prima, and T. meandrinus are lamellate. Thorecta reticulata is massive, with upright turrets and without a peduncle. The new species is closest to T. latus and T. tuberculatus by its massive globular, stalked growth form. Thorecta atlantica sp. nov. can be easily distinguished from both species by its thicker fibres and very large apical oscules, limited to 1–2 per specimen, whereas T. latus and T. tuberculatus have numerous small oscules (Carter, Reference Carter1885).
The species composition of the genus Thorecta has been subject of a great deal of controversy since its creation, ranging from the original 24 species (Lendenfeld, Reference Lendenfeld1889) to 9 (de Laubenfels, Reference De Laubenfels1948), 12 (Cook & Bergquist, Reference Cook and Bergquist1996) and 20 species (Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002). This high variation is due to the wide range of concepts of the genus and species by different authors, as well as by misinterpretations of taxonomic characters (Cook, Reference Cook, Custódio, Lôbo-Hajdu, Hajdu and Muricy2007). Lendenfeld (Reference Lendenfeld1888, Reference Lendenfeld1889) created the genus to include ‘Spongiidae with a loose skeleton-net, with meshes 0.5–1.2 mm wide; stout, simple or branched connecting fibres, and a thick sand cortex on the smooth surface. Without superficially extending oscular tubes and corresponding grooves in the skeleton’. No mention was made to the presence or absence of foreign detritus in either primary or secondary fibres or to the stratification of the fibres; Lendenfeld thus included in Thorecta species with cored primary and secondary fibres, which would be currently assigned to Hyrtios Duchassaing & Michelloti, Reference Duchassaing de Fonbressin and Michelotti1864 (e.g. T. laxus Lendenfeld, Reference Lendenfeld1889; T. madagascarensis Lendenfeld, Reference Lendenfeld1889).
De Laubenfels (Reference De Laubenfels1948) stressed the importance of the stratification of spongin fibres, with cored primary fibres and uncored secondary fibres, since these characters were cited in the original description of the type species T. exemplum var. tertia (Lendenfeld, Reference Lendenfeld1888, Reference Lendenfeld1889). He amended the definition of the genus to include Aplysinopsis Lendenfeld, Reference Lendenfeld1888 (which has conules and thinner dermal cortex) and Thorectandra Lendenfeld, Reference Lendenfeld1889 (which has larger meshes and high amount of mucus). De Laubenfels (Reference De Laubenfels1948) also included species with cored secondary fibres (e.g. Alcyonium boletus Lamarck, Reference Lamarck and De Monet1815; Aplysinopsis elegans Lendenfeld, Reference Lendenfeld1888). His extensive list of synonyms, however, reduced the genus to nine valid species: T. boleta, T. farlovi (Hyatt, Reference Hyatt1877 as Spongelia), T. pala (Lamarck, Reference Lamarck1814 as Spongia), T. byssoides (Lamarck, Reference Lamarck1814 as Spongia), T. calyx (Lamarck, Reference Lamarck1814 as Spongia), T. murrayi (Poléjaeff, Reference Poléjaeff1884 as Cacospongia), T. variabilis (Poléjaeff, Reference Poléjaeff1884 as Luffaria), T. elegans (Lendenfeld, Reference Lendenfeld1888 as Aplysinopsis) and T. penicillosa (Lamarck, Reference Lamarck1814 as Spongia).
Wiedenmayer (Reference Wiedenmayer1977) added Thorecta horridus (Hyatt, Reference Hyatt1877 as Hircinia cartilaginosa var. horrida), and later (Wiedenmayer, Reference Wiedenmayer1989) included T. choanoides (Bowerbank, Reference Bowerbank1872 as Halispongia, now considered as a Thorectandra; see Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994), T. tuberculatus (Carter, Reference Carter1885 as Stelospongus) and T. glomerosus Wiedenmayer, Reference Wiedenmayer1989. Bergquist (Reference Bergquist1980) apparently considered valid only nine species of Thorecta: T. murrayi, T. meandrinus, T. latus, T. calyx, T. madagascarensis, T. byssoides, T. vasiformis and T. freija; Thorecta pala was listed by Bergquist (Reference Bergquist1980) as valid in p. 470, but was included in Taonura in p. 472. Eight species described by Lendenfeld (Reference Lendenfeld1888, Reference Lendenfeld1889) were considered unrecognizable by Bergquist (Reference Bergquist1980): T. pumilus, T cacos, T. murrayella, T. exemplum var. prima, T. laxus, T. crateriformis, T. dendroides, and T. tuberculatus.
Hooper & Wiedenmayer (Reference Hooper, Wiedenmayer and Wells1994) included 12 Australian species in their catalogue: T. calyx, T. donar Lendenfeld, Reference Lendenfeld1889, T. farlovi, T. freija Lendenfeld, Reference Lendenfeld1889, T. latus (Carter, Reference Carter1885 as Stelospongus), T. meandrinus Lendenfeld, Reference Lendenfeld1889, T. murrayi, T. polygonum (Lendenfeld, Reference Lendenfeld1889 as Coscinoderma), T. prima Lendenfeld, Reference Lendenfeld1888, T. tuberculatus, T. typhinus (Lamarck, Reference Lamarck1814), and T. vasiformis (Carter, Reference Carter1885 as Geelongia). On the other hand, they transferred T. boletus, T. choanoides and T. glomerulosus to Thorectandra. Cook & Bergquist (Reference Cook and Bergquist1996) considered 12 species as valid: T. byssoides, T. calyx, T. farlovi, T. horridus, T. murrayi, T. latus, T. vasiformis, T. freija, T. madagascarensis, T. meandrinus, T. cf. polygonum and T. reticulata Cook & Bergquist, Reference Cook and Bergquist1996. In the most recent revision of the genus, Cook & Bergquist (Reference Cook, Bergquist, Hooper and Van Soest2002) stated that it includes 20 species, without naming them.
A detailed revision of all species related to Thorecta, including re-examination of all type specimens is greatly needed but is outside the scope of this paper. We tried, however, to reassign these species names according to the currently accepted definitions of Thorectid genera (Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002; Cook, Reference Cook, Custódio, Lôbo-Hajdu, Hajdu and Muricy2007) and to the synonyms suggested in the literature, in order to allow comparisons with the new Brazilian species. Our analysis indicates that many of these species should be relocated to different genera. Thorecta choanoides is probably well placed in Thorectandra due to its large mesh size and production of mucus (Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994; Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002). Among the species included in Thorecta by Hooper & Wiedenmayer (Reference Hooper, Wiedenmayer and Wells1994), T. calyx, T. freija, T. murrayella and T. typhina differ from Thorecta sensu Cook & Bergquist, Reference Cook, Bergquist, Hooper and Van Soest2002 by the absence of a dermal armour, and would be better classified in Taonura (stalked species) and Cacospongia (massive species). Thorecta galeiformis and T. murrayi have tertiary fibres and should thus be included in Luffariella or Phyllospongia. Cacospongia exemplum var. secunda (Hooper & Wiedenmayer, Reference Hooper, Wiedenmayer and Wells1994; as a junior synonym of T. latus) has uncored primary and secondary fibres and is therefore a Datcylospongia. Thorecta madagascarensis (referred as a junior synonym of T. byssoides) is unarmoured and has both primary and secondary fibres cored by detritus; it probably should be included in Hyrtios, although the presence of a stalk is atypical in that genus.
The genus Thorecta should probably be left with only 11 species, which have all the diagnostic characters of the genus (Table 2): T. carteri, T. marginalis, T. primus, T. latus, T. farlovi, T. meandrinus, T. polygonum, T. reticulatus, T. tuberculatus, T. vasiformis, and T. atlantica sp. nov. Wiedenmayer (Reference Wiedenmayer1977: 70) suggested that Thorecta should be treated as a masculine name because it was not derived from either a Latin or a Greek word, but is instead an arbitrary latinization of the Greek Thorectes (=armoured warrior). The gender of Thorecta, however, was not specified by Lendenfeld (Reference Lendenfeld1888) and cannot be deduced from the names of the species originally included, which are variously masculine, feminine and neuter. In this case article 30.2.4 of the ICZN (1999), which states that if such name ends in ‘-a’ the gender is feminine, should be applied. Therefore, we consider that Thorecta should be treated as a feminine name. In consequence, the specific epithets of some of the species of the genus should be changed to the feminine to agree in gender with the genus name (ICZN, 1999: article 31.2): T. latus to T. lata; T. meandrinus to T. meandrina; T. polygonum to T. polygona; T. reticulatus to T. reticulata; and T. tuberculatus to T. tuberculata. The names finishing in ‘-is’, such as marginalis and vasiformis, can be both feminine and masculine; together with the names carteri and farlovi, derived from personal names, they need not to be changed.
Table 2. Main characters of valid species of Thorecta. F1, cored primary fibres; F2, uncored secondary fibres. All measurements are width in μm.
S, South; W, Western; E, Eastern; N, Northern; *References: 1, Bowerbank (Reference Bowerbank1872); 2, Hyatt (Reference Hyatt1877); 3, Carter (Reference Carter1885); 4, Lendenfeld (Reference Lendenfeld1885); 5, Lendenfeld (Reference Lendenfeld1888); 6, Lendenfeld (Reference Lendenfeld1889); 7, de Laubenfels (Reference De Laubenfels1948); 8, Wiedenmayer (Reference Wiedenmayer1989); 9, Hooper & Wiedenmayer (Reference Hooper, Wiedenmayer and Wells1994); 10, Cook & Bergquist (Reference Cook and Bergquist1996); 11, Cook & Bergquist (Reference Cook, Bergquist, Hooper and Van Soest2002).
The tortuous history of the genus Thorecta, its controversial scope and the disputed synonyms of many species are examples of the problems that poorly-described species can cause in sponge classification. Different authors have made great efforts to interpret such poor descriptions; the lack of information and their diverging views on the importance of characters such as growth form, surface sculpturing, dermal armour, fibre stratification, fibre coring and presence of tertiary fibres led to the currently confused classification of the genus. Detailed descriptions of fresh material, well illustrated by photographs and including as many characters as possible, are essential tools for the clarification of the systematics of Thorecta in the future.
ACKNOWLEDGEMENTS
This work was funded by grants and fellowships provided by CENPES/PETROBRAS, CAPES, CNPq, FAPERJ and FACEPE.
