Cercicladia australis, a new carnivorous sponge with novel chelae from the Tasman Basin and the Argentine Patagonian Margin (Porifera, Cladorhizidae) Author Rios, Pilar Author Kelly, Michelle Author Vacelet, Jean text Zootaxa 2011 3131 52 62 journal article 45784 10.5281/zenodo.205649 0692e4e9-6d53-43fb-a934-81ee679b25a7 1175-5326 205649 Cercicladia australis gen. nov. sp. nov. ( Fig. 2–5 ) Material examined. Holotype— NIWA 39599, NIWA Stn TAN0803/19, Seamount 1, Macquarie Ridge, New Zealand , 48.532°S , 164.947°E , 1060–1112 m , epibenthic sled, 30 Mar 2008 . Paratypes ― NIWA 39631, NIWA Stn TAN0803/21, Seamount 1, Macquarie Ridge, New Zealand , 48.526°S , 164.941°E , 1096–1364 m , epibenthic sled, 30 Mar 2008 ; MNCN 1.01/640, RV Miguel Oliver Stn BC5, off east coast of Patagonia, 44.769ºS , 59.439ºW , 1728 m , box corer, 9 Feb 2009 ; MNHN DCL 4096, RV Miguel Oliver Stn DR_4, off east coast of Patagonia, 46.959ºS , 59.396ºW , 1225 m , rock dredge, 13 Nov 2008 . Other material. Macquarie Ridge , New Zealand , Seamount 3 : NIWA 39774, NIWA Stn TAN0803/33, 50.091°S , 163.482°E , 1077–1408 m , 1 Apr 2008 (spicules of C . australis gen. nov. sp. nov. contaminate this specimen of Echinostylinos stylophora (Lévi & Lévi, 1983) ; NIWA Stn TAN0803/32 ( DTIS image Fig. 2 B), 50.0952°S , 163.461°E , 1697– 1091 m , 1 Apr 2008 . Seamount 1 : NIWA Stn TAN0803/20 ( DTIS image Fig. 2 A), 48.558°S , 164.939°E , 1106–1209 m , 30 Mar 2008 . East coast of Patagonia : MNCN 1.01/641– MNCN 1.01/649, RV Miguel Oliver Stn DR15, 45.414ºS , 59.206ºW 1577 m , rock dredge, 26 Jan 2008 ; MNCN 1.01/650: RV Miguel Oliver Stn DR4, 46.468ºS , 59.377ºW , 1145-1407 m , rock dredge, 10 Apr 2008 ; MNCN 1.01/651: RV Miguel Oliver Stn DR4, 46.468ºS , 59.377ºW , 1145–1407 m , rock dredge, 13 Nov 2008 . Type locality. Seamount 1, Macquarie Ridge, New Zealand . Distribution. Macquarie Ridge, southern New Zealand , and the southwest Argentine margin of Patagonia. Description. Erect, feather-shaped sponge with a thick fleshy pedicle, inflated near the base, becoming progressively thinner above and ending in a flattened blade that bears biserially arranged lateral filaments of various lengths ( Fig. 2 A–C, 3A–L). Holotype (NIWA 39599) 77 mm total length (pedicle intact, 62 mm long, 5.3 mm diameter at base, 1.2 mm diameter at apex; blade damaged, 12 mm long; Fig. 2 C left). Paratype NIWA 39631 72 mm total length (pedicle absent, blade damaged, width about 10 mm , filaments broken, 5 mm long, 0.6 mm thick; Fig. 2 C right). Epibenthic sled images of the sponges in life ( Fig. 2 A, B) indicate that width of the undamaged blade may be at least 30 mm . Surface granular. Texture of pedicle is fleshy; texture of blade is tough, filaments are firm. In two of the Patagonian specimens, the pedicle is covered with a feltwork of acanthosubtylostyles, sediment and foreign spicules ( Fig. 3 M–Q, 4B–C). The color in life is translucent creamy white, translucent brown to whitish in alcohol. The sponge contains numerous shells of foraminifera. There is no visible aperture or canal system. Skeleton. Pedicle and blade axes composed of long mycalostyles arranged longitudinally in the blade, twisted in the pedicle, and packed tightly overall. Styles in the center of the shaft and in the filaments are generally thinner ( Fig. 3 M, N) and anchored by their proximal end in the centre of the shaft, projecting outwards ( Fig. 3 M). Living tissue poorly preserved, forming a thin layer around the axes and bases of the filaments, containing a huge number of randomly dispersed cercichelae. The microscleres are scattered throughout the sponge, although they are more abundant near the apex. Acanthosubtylostyles, when present, form a feltwork on the pedicle ( Fig. 4 B–D). Megascleres ( Table 1 )― Mycalostyles ( Fig. 2 D, 5A), straight, fusiform, with a short, acerate point and a short subtylostylote restriction near the proximal end; 775–2504 Μm long x 18–84 Μm thick. Acanthosubtylostyles ( Fig. 5 B) in a broad size range were found in three specimens from Patagonia. These spicules were thin, microspined, the proximal end being subtylote to occasionally tylote, the distal end having a blunt point; 84 (50–125) m long, x 2 (1–4) Μm thick (MNCN 1.01/643), 93 (52–130) Μm long, x 2 (1–4) Μm thick (MNCN 1.01/651), and 285 (65– 482) Μm long, x 2 (1–4) Μm thick (MNCN 1.01/640). Microscleres ( Table 1 )― Microxea ( Fig. 2 E, 5C), thickened centrally with attenuated and faintly curved ends, slightly roughened, very numerous in Macquarie Ridge specimens, but uncommon in some Patagonian specimens, 97–223 Μm long x 1–7 Μm thick. Toxas ( Fig. 2 F, 5E), faintly roughened, slightly or deeply curved, often with a slight swelling in the middle, uncommon, possibly absent in some specimens, but very likely not foreign, 15–92 Μm long x 0.5 Μm thick. Sigmancistras ( Fig. 2 G, 5F), symmetrical, with a broadened shaft and thin, curved tips aligned in the same plane, with very faint notches within the curved tip, and occasionally with a marked central notch, C- or S-shaped, extremely common, 28–54 Μm long x 2–6 Μm thick. Cercichelae ( Fig. 2 H, 5D), in the shape of a narrow, elongate, oval ring, with the two sides slightly unequal in thickness, the cavity displays a central inflexion, and is often in the shape of a laterally flattened rhombus or oval. Short, conical spines are present on the external side of the ring. These spines are irregularly dispersed, although generally more numerous near the two ends of the spicules, and absent in the middle. In face view ( Fig. 5 D lower left) upper and lower teeth align centrally, eventually fusing; spines are concentrated centrally on opposite ends. In a few cases, most probably in immature spicules, one side is incompletely closed, displaying a suture line ( Fig. 2 H lower middle and 5D upper right), or unfused teeth ( Fig. 2 H lower right and 5D lower), suggesting that the microsclere originates from a chela; Macquarie Ridge 82–92 µm, Patagonia 52–78 µm. Substrate, depth range, and ecology. Seamounts 1 and 3 are elevated features on the Macquarie Ridge, reaching 500 m and 550 m below sea-level, respectively ( Rowden et al . 2008 ). The bottom substrate of both seamounts is dominated by disarticulated barnacle plates, bedrock, gravel, and sand. The faunal groups most commonly observed on the seabed DTIS images were sponges and sea pens, but asteroids, hydrocorals, gorgonian corals and echinoids were also present. Specimens were collected on the Macquarie Ridge seamounts between 1060 and 1408 m. Patagonian specimens were collected from sandy substrate, mostly fine medium sand with low organic content, between 1145 and 1728 m. TABLE 1. Cercicladia australis gen. nov. sp. nov. , megasclere and microsclere dimensions (Μm). The number of measurements, if stated, is given as n=x. The term ‘rare’ refers to data that is based on less than 10 measurements. Mycalostyles Acanthosubtylostyles Microxeas Toxas Cercichelae Sigmancistras NIWA 39599 900–1700 (1305 ± 325) absent 110–200 (162 ± 30) 40– 48 x 1, rare 85–91 (88 ± 2) 28–46 (37 ± 6) x 4–5 (4 ± 1) Holotype x 28–48 (37 ± 6) x 2–5 (3 ± 1) NIWA 39631 1250–2500 (1890 ± 441) absent 150–200 (166 ± 21) absent 82–92 (88 ± 4) 35–54 (45 ± 6) x 4–5 (5 ± 1) Paratype x 30–51 (43 ± 7) x 2–5 (4 ± 1) MNCN 1.01/640 1225–2100 (1653 ± 229) 62–482 (285±76) 120–188 (158 ± 16) 25–58 (42 ± 6) 58–68 (62 ± 3) x 10–15 (14 ± 2) 28–40 (34 ± 3) x 3–5 (3 ± 1) Paratype x 25–43 (34 ± 4) x 1–4 , n=150 x 3–5 (5 ± 1) MNHN DCL4096 1299–2264 (1836 ± 250) absent 118–213 (194± 19) absent 52–77 (64 ± 6) x 10–15 (13 ± 1) 29–40 (33 ± 3) x 3–5 (4) Paratype x 18–84 (39 ± 11) x 2–7 (5 ± 1) MNCN 1 . 01/641 775–2287 (1549 ± 452) absent 110–210 (167 ± 29) 18–63 (34 ± 13) 58–70 (63 ± 4) x 13–20 (15 ± 2) 28–45 (35 ± 5) x 3–5 (3 ± 3) x 30–45 (37± 5) x 1–6 (4 ± 1) MNCN 1 . 01/642 1046–2351 (1046 ± 340) absent 114–222 (169 ± 33) 26–63 (34 ± 11) 57–73 (64 ± 4) x 12–16 (15 ±1) 29–48 (37 ± 4) x 2–5 (4 ± 1) x 21–54 (33 ± 7) x 2–7 (4 ± 2) MNCN 1 . 01/643 1000–2100 (1738 ± 320) 50–125 (84 ± 20) 130–210 (148 ± 26) 23–93 (44 ± 23), rare 55–73 (65 ± 7) x 10–18 (15 ± 1) 28–43 (34 ± 4) x 3–4 (3 ± 1) x 28–40 (36 ± 5) x 1–4 (2 ± 4), n= 14 x 3–6 (4 ± 1) MNCN 1 . 01/644 1200–2325 (1670 ± 278) absent 108–190 (138 ± 22) 25–53 (41 ± 8) 55–70 (64 ± 4) x 13–15 (14 ± 1) 30–43 (34 ± 4) x 3–5 (3 ± 1) x 25–48 (34 ±6) x 1–5 (2 ± 1) MNCN 1 . 01/645 1146–2414 (1882 ± 288) absent 115–223 (165 ± 35) 19–53 (36 ± 10) 57–71 (63 ± 3) x 12–17 (14 ± 1) 27–44 (35 ± 5) x 2–6 (4 ± 1) x 24–41 (32 ± 5) x 1–7 (4 ± 2) MNCN 1 . 01/646 1225–2250 (1795 ± 283) absent 98–209 (153 ± 38) absent 58–75 (66 ± 4) x 10–18 (14 ± 2) 29–43 (35 ± 4) x 3–5 (3 ± 1) x 23–48 (38 ± 6) x 1–6 (3 ± 2) MNCN 1 . 01/647 1092–2433 (1878 ± 357) absent 115–212 (179 ± 25) absent 58–74 (66 ± 4) x 13–17 (15 ± 1) 28–47 (36 ± 5) x 3–5 (4 ± 1) x 19–45 (32 ± 7) x 3–7 (5 ± 1) MNCN 1 . 01/648 1354–2232 (1872 ± 250) absent 97–217 (168 ± 30) 21–49 (37 ± 9) 55–77 (64 ± 5) x 11–21 (15 ± 2) 30–43 (35 ± 3) x 3–6 (4 ± 1) x 24–46 (37 ± 6) x 1–7 (4 ± 2) MNCN 1 . 01/649 933–1504 (1323 ± 146) absent 115 x 3 , n= 1 18–52 (30 ± 19), rare 54–78 (66 ± 6) x 12–17 (14 ± 1) 28–42 (34 ± 4) x 2–5 (4 ± 1) x 21–52 (33 ± 7) MNCN 1 . 01/650 1100–2275 (1738 ± 272) absent 98–200 (157 ± 31) 15–45 (35 ± 9) 58–75 (64 ± 5) x 10–18 (14 ± 2) 28–43 (35 ± 4) x 3–5 (4 ± 1) x 23–38 (30 ± 4) x 1–5 (31 ± 2) MNCN 1 . 01/651 913–2504 (1649 ± 381) 52–130 (93 ± 23) 102–186 (143 ± 29), n=6 37–49 (41 ± 5), rare 52–75 (62 ± 5) x 12–16 (14 ± 1) 29–40 (34 ± 3) x 2–5 (3 ± 1) x 20–44 (32 ± 7) x 1–4 (2 ± 1), n=12 FIGURE 2 . Macquarie Ridge Cercicladia australis gen. nov. sp. nov. : A . in situ DTIS image from RV Tangaroa Stn TAN0803/020, Macquarie Ridge, 1106–1209 m, blade width of sponge about 25 mm; B . in situ DTIS image from RV Tangaroa Stn TAN0803/032, Macquarie Ridge, 1091–1697 m, blade width of sponge about 30 mm; C . holotype NIWA 39599 (left), paratype NIWA 39631 (right); D . proximal and distal ends of a mycalostyle; E . microxeas; F . toxa; G . sigmancistras; H . cercichelae at various stages of development, from fully fused (upper) to separated teeth (lower right) (D–H SEM images from the holotype NIWA 39599). FIGURE 3 . Patagonia Cercicladia australis gen. nov. sp. nov. : A–L ex situ images of paratypes and other specimens; A . paratype MNCN1.01/640, pedicle base (left), remainder of pedicle and blade (right); B . paratype MNHN DCL 4096; C . MNCN 1.01/642; D . MNCN 1.01/648; E . MNCN 1.01/645; F . MNCN 1.01/649; G . MNCN 1.01/643; H . MNCN 1.01/646; I . MNCN 1.01/641; J . MNCN 1.01/651; K . MNCN 1.01/644; L . MNCN 1.01/647; M-Q SEM images of histological details of the blade and filaments; M . showing locations of magnified views in O, N, Q, upper box indicating position of magnified view in O, middle box indicating position of magnified view in N, and lower box indicating position of magnified view in Q; N . primary and filament axes composed of longitudinally arranged, tightly packed, mycalostyles; O . cercichelae and sigmancistras in mesohyl connecting primary and lateral filaments; P . sigmancistras resting on megascleres; Q . sigmancistras in mesohyl connecting tracts within the primary axis (M–Q from MNCN 1.01/641). FIGURE 4 . Patagonia Cercicladia australis gen. nov. sp. nov. , skeleton of paratype MNCN1.01/640: A . plan of overall morphology; B . external surface showing felted acanthosubtylostyles; C . detailed view of felted acanthosubtylostyles; D . acanthosubtylostyles; E . sigmancistras resting on mycalostyles of primary axis. Etymology. Named for the southern New Zealand and Atlantic Ocean locations from whence this species was discovered ( australis , Latin for ‘southern’). Remarks. Cercicladia australis gen. nov. sp. nov. has the general morphology of many deep sea carnivorous sponges, being pinnate with a long stem. The pedicle is unusual in that it is fleshy, and increases in diameter towards the base. DTIS images of the species in situ on the Macquarie Ridge ( Fig. 2 A, B) indicate that the sponge has long, thin filaments disposed in a regular series along two sides of a flattened blade with the overall blade and filament forming a feather–like structure that is curved in the broader plane. The filaments are easily broken upon collection. Although direct evidence of the presence of partly digested prey was not found in or on these specimens, their morphology strongly suggests a carnivorous feeding habit, and thus affinity with one of the families of carnivorous poecilosclerids, predominantly Cladorhizidae , but also including Guitarridae , and Esperiopsidae ( Hajdu & Vacelet 2002 ) . The species has the typical morphology and mycalostyles of all carnivorous sponges. It also has sigmancistras, known thus far only in carnivorous sponges, and present in at least some representatives of all genera of Cladorhizidae , and in Euchelipluma Topsent, 1909 (Guitarridae) . The acanthosubtylostyles of the stalk seem to be similar to those found in some species of Asbestopluma Topsent, 1901 , but it lacks the true chelae found in all other carnivorous sponges, having only a highly modified derivative, the cercichelae. FIGURE 5 . Patagonia Cercicladia australis gen. nov. sp. nov. , spicules from paratype MNCN1.01/640: A . proximal and distal ends of a mycalostyle; B . acanthosubtylostyle (left), proximal and distal ends (right); C . microxea; D . cercichelae at various stages of development, from fully fused to separated teeth (lower right); E . toxa; F . sigmancistras. The cercichelae are somewhat puzzling as they are very similar in shape to the canonochelae of Isodictya lankesteri , first described as Cercidochela lankesteri from Antarctica , but the resemblance is only superficial. The canonochela appears to be derived from a chela with lateral alae or lamellae, as does the cercichelae of Cercicladia australis gen. nov. sp. nov. , but the latter has no lateral alae, only the suggestion of a frontal tooth in the immature form. Many of the cercichelae are also asymmetrical longitudinally, with what appears to be a restriction on one side of the oval loop ( Fig. 2 H upper right, lower middle, 5D upper right). A few spicules have been observed with incompletely fused ends on one side ( Fig. 2 H lower middle and right and 5D lower). This form may be somewhat similar to the clavidisc microsclere of Merlia Kirkpatrick, 1908 ( Hoshino 1990 ) , which, however, derives from diancistra rather than from a chela. The other microscleres of the sponge are also rather puzzling in the context of the family Cladorhizidae . The sigmancistras are broad longitudinally with incurved tips, a feature characteristic of the sigmancistras of most cladorhizid sponges, but they also have the faint suggestion of a notch in the upper and lower curves, which indicates the nature of true sigmancistras. A more striking peculiarity is the presence of numerous highly attenuated microxeas, and of rare bow–shaped microxeas that we consider as toxas in other demosponges. The new genus Cercicladia gen. nov. is hereby designated as a member of the Family Cladorhizidae on the basis of the possession of the feather–like morphology, the lack of evidence of an aquiferous system, the possession of mycalostyles, acanthosubtylostyles (albeit, only in a few specimens) and sigmancistras, even though it lacks the (an)isochelae considered to be diagnostic of the Family Cladorhizidae . The microxea and toxa have not been reported in other carnivorous sponges thus far, although they frequently occur in the suborder Mycalina in which Cladorhizidae are classified (Hajdu, Van Soest & Hooper 1994 ). However, similar microxeas and toxa–like microscleres are known from another carnivorous sponge with chelae, presently being described from the same location in New Zealand . While the possession of toxa in Cercicladia gen. nov. strengthens the affinity of Family Cladorhizidae with Mycalidae , the discovery of cercichelae and microxea in this new genus does not, unfortunately, improve our understanding of the affinity of Cladorhizidae with other groups. The distribution of Cercicladia australis gen. nov. sp. nov. is remarkable in that it has been found in only two subpolar locations on either side of the globe, on the Macquarie Ridge to the southwest of New Zealand , and off the Argentine coast of Patagonia. We are confident about this disjunct distribution because the diagnostic spicules are unique and very easily recognizable, and considerable collection efforts in both locations have been made over many years ( Gordon et al . 2010 ). Although several carnivorous sponges appear to have broad distributions in the deep sea, (for instance Cladorhiza abyssicola Sars, 1872 ), these may be artificial due to the difficulty of identification of species in which the various spicules are differentially distributed around the body of the sponge, and which do not have special diagnostic spicules. On the contrary, most highly recognisable species such as Asbestopluma ( Asbestopluma ) anisoplacochela Kelly & Vacelet, 2011 , A . ( A .) desmophora Kelly & Vacelet, 2011 , and Abyssocladia carcharias Kelly & Vacelet, 2011 , seem to have relatively restricted distributions. In the case of C . australis gen. nov. sp. nov. , the spicule complement is unique enough to rule out the hypothesis of approximate identification, but it may be hypothesized that we are dealing with two vicariant species. The specimens at hand from the Macquarie Ridge and the Argentine margin of Patagonia, however, are surprisingly similar, and the slight differences in spicule size and complement are not considered of great significance. Thus, we consider them as a single species.