Pseudosuberites thurberi sp. nov.

( Figs 1, 9–12; Table 6)

Pseudosuberites sp., Baco et al., 2010: 255, 256, fig 2b.

Pseudosuberites sp., Thurber et al., 2010: 260, 265–268, fig. 3, table 3.

Pseudosuberites sp., Bowden et al., 2013: 5 –11.

Material examined. South and North Tower Seeps, Opouawe Bank, Hikurangi Margin: Holotype — NIWA 27044, NIWA Stn TAN0616/79, North Tower, 41.783° S, 177.399° E, 1040–1053 m, collected by epibenthic sled, 13 Nov 2006. Paratypes — NIWA 27043, 27045, 27047, 27060, NIWA Stn TAN0616/79, North Tower, 41.783° S, 177.399° E, 1040–1053 m, collected by epibenthic sled, 13 Nov 2006; NIWA 27055, NIWA Stn TAN0616/83, South Tower, 41.783° S, 175.399° E, 1050–1053 m, collected by epibenthic sled, 13 Nov 2006.

LM-3 Seep, Rock Garden, Hikurangi Margin: NIWA 32043, IFM GEOMAR Stn SO 191-3/238, 39.977° S, 178.236° E, 907–908 m, collected by TV grab, 6 Mar 2007.

LM-9 Seep, Omakere Ridge, Hikurangi Margin: NIWA 32063, IFM GEOMAR Stn SO 191-2/164, 40.054° S, 177.822° E, 1097–1110 m, collected by TV grab, 22 Feb 2007.

Other material. South and North Tower Seeps, Opouawe Bank, Hikurangi Margin: NIWA 27046, 27048, NIWA Stn TAN0616/79, North Tower, 41.783° S, 177.399° E, 1040–1053 m, collected by epibenthic sled, 13 Nov 2006; NIWA 27051, NIWA Stn TAN0616/80, South Tower, 41.790° S, 175.405° E, 1055– 1050 m, collected by epibenthic sled, 13 Nov 2006; NIWA 27052, 27054, NIWA Stn TAN0616/83, South Tower, 41.783° S, 175.399° E, 1053– 1050 m, collected by epibenthic sled, 13 Nov 2006; NIWA 35004, IFM GEOMAR Stn SO 191-2/149, South Tower, 41.789° S, 175.407° E, 1055 m, collected by TV grab, 19 Feb 2007.

LM-9 Seep, Omakere Ridge, Hikurangi Margin: NIWA 27034, NIWA Stn TAN0616/45, 40.013° S, 177.860° E, 1127–1160 m, collected by epibenthic sled, 7 Nov 2006; NIWA 32040, 32061, 35008, IFM GEOMAR Stn SO191-2/87, 40.053° S, 177.735° E, 1106– 1098 m, collected by TV grab, 12 Feb 2007; NIWA 32062, IFM GEOMAR Stn SO191-2/164, 40.054° S, 177.822° E, 1097–1110 m, collected by TV grab, 22 Feb 2007.

Type location. South and North Tower Seeps, Opouawe Bank, Hikurangi Margin.

Distribution. Opouawe Bank at southern entrance to Cook Strait, south of Wairarapa coast, North Island, 907– 1160 m, north to Omakere and Ritchie’s Ridge on the Hikurangi Margin.

Description. Thickly encrusting, widely spreading, mounded sponge, giving rise to multiple, finely divided branches, that form a highly fragmented mass ( Figs 9, 10) about 4–5 cm high and wide. Sponges on hard substrate tend to form encrusting mounds, while those partially buried in sediment produce fine tendrils ( Figs. 9, 10B). Branches may be relatively thick with incipient branching, up to 1 cm diameter (e.g. Figs 10A, C), to finely divided tendrils 2–3 mm diameter (e.g. Fig. 10B). Oscules are not visible in life or in the preserved state. Texture is corky, slightly granular to the touch, overall firm, compressible, and flexible. Colour in life, dull greyish white; in preservative, dull tan to light orange brown.

Skeleton. Choanosome composed of loose, divaricating bundles of tylostyles, aligned longitudinally in surface projections, running parallel with the base of the sponge in encrustations ( Fig. 11A), ranging from about 100–150 µm wide. Free spicules are interspersed between tracts, criss-crossing the primary tracts as single spicules or forming deep, diverging, subectosomal brushes ( Figs 11B, C). Tracts or subectosomal brushes emerge at the surface as a dense palisade of bouquets of tylostyles, through which lies a relatively thin layer of predominantly, tangentially disposed tylostyles. The whole forms a dense, thick, continuous palisade.

Spicules. Megascleres ( Table 6; Fig. 12), tylostyles with well-developed, spherical heads, 291 (134–530) µm long × 8 (7–15) µm thick, with slight protrusion of the apex, shaft uniform along the length and slightly bent off centre in the upper third. No obvious difference in the lengths of the tylostyles between the ectosome and choanosome.

Substrate, depth range and ecology. Attached to authigenic carbonate rock, 907–1160 m. Thurber et al. (2010) hypothesise that this species may be chemoautotrophic and play a significant role in facilitating the transfer of methane into the metazoan food web in hard substrata habitats at the seep sites.

Etymology. The species is named in honour of Dr Andrew Thurber, Department of Microbiology, Oregon State University, U.S. A, for his leadership in the discovery and first description of a unique sponge-dominated community that is largely fuelled by methane from the seeps ( Thurber et al. 2010).

Remarks. A general survey of presently accepted species in Pseudosuberites ( Van Soest et al. 2018b) reveals a range of choanosomal architectures, seemingly dependent upon whether the species is encrusting (confused skeleton with loose tracts) or lobate/digitate (some axial compression of tracts with subectosomal divergence towards ectosomal bouquets), indicating a progression in tract and axial compression in the skeleton as the sponge thickens and attains height. Megascleres in the choanosomal skeleton also vary considerably in their overall length, ranging from about 350 µm in P. andrewsi Kirkpatrick and P. exalbicans Topsent, to 1200 µm in P. hyalinus and 2000 µm in P. nudus Koltun.

Four species are known from the northern hemisphere: P. montiniger (Carter) from the North and East Barents Sea; P. sadko Koltun from the Russian Arctic Ocean; P. mollis Topsent from the Western Mediterranean; P. sulphureus, a thin encrusting sponge from the North Sea. Eight species are known from the Indo-Pacific: three from Japan [( P. incrustans (Thiele), P. kunisakiensis Hoshino, and P. perforates (Thiele)], one from Easter Island ( P. vakai Desqueyroux-Faundez) and one from the Marshall Islands [( P. purpureus (de Laubenfels)]. Pseudosuberites andrewsi Kirkpatrick, is perhaps the best known tropical species, from Christmas Island in the Western Indian Ocean; P. cava Sollas was recorded from the Malay Peninsula, and P. lobulatus (Lévi) from Vietnam. However, most species have been recorded from Antarctica and the southern-most reaches of the South Pacific Ocean ( Chile and the Patagonian Shelf), and the South Atlantic Ocean ( Tristan Gough) ( Table 7).

Bergquist (1968: 24–26) ascribed several sponges from Auckland’s Waitemata Harbour, and the Hauraki Gulf’s North Channel, to Pseudosuberites sulcatus ( Thiele, 1905) which was first described from waters off Cabo de Espiritu Santo ( 53.783° S, 67.500° W) on the Patagonian Shelf in the South Atlantic Ocean ( Table 7). Bergquist’s specimens are similar to P. sulcatus, as originally described by Thiele (1905) and Bergquist (1968), including the spicule dimensions [Cabo de Espiritu Santo: choanosomal tylostyles 370 µm long, 12 µm thick, ectosomal tylostyles 175 µm long, 5 µm thick; Narrow Neck: tylostyles 284 (170–386) µm long, 9.6 (6.0–12.7) µm thick; North Channel tylostyles 272 (140–387) µm long and 7.3 (4.6–8.4) µm thick]. The sponges are small and thinly encrusting [Cabo de Espiritu Santo: about 15 mm high with surface processes; New Zealand: 0.8–1.4 mm thick with irregular surface processes (2.8–4.0 mm high, 1.2–1.6 mm wide)], both with a ‘choanosome composed of a confused mat of large tylostyles with a tendency towards vertical disposition, and the ectosome was composed of vertically arranged tylostyles of large and small size, with a tendency for small spicules to predominate in the dermal region ( Bergquist 1968). Thiele (1905) described some ‘bundle’ formation in the choanosome, tylostyles being bound by spongin, and the same ‘radial’ arrangement of spicules in the ectosome.

What is quite clear from these descriptions and recent descriptions of P. sulcatus from Easter Island ( Desqueyroux-Faúndez, 1990), is that P. sulcatus (and the New Zealand specimens assigned to this species) does not conform to our current understanding of the genus Pseudosuberites, i.e. being the only suberitid genus with an ectosome of tangentially orientated megascleres of a comparable size to those in the choanosome. Rather, P. sulcatus (and the New Zealand specimens) conform more closely to the genus Protosuberites, as diagnosed by Van Soest (2002) and recently revisited in Van Soest & Kluijver (2003) and Samaai & Gibbons (2005), as possessing a ‘surface skeleton of brushes of tylostyles, which are often somewhat smaller than those of the choanosome. Choanosomal skeleton consists variably of single spicules erect on the substrate or bundles running from the substrate to the surface, usually parallel to each other, without any form of anastomosing’ ( Van Soest 2002: 235).

Bergquist (1968) followed Burton (1930) in also relegating the New Zealand species Suberites ramosus, S. anastomosus and S. incrustans, in synonymy with P. sulcatus. These three species were first described by Brøndsted (1924) from the shallow subtidal in Perseverance Harbour on Campbell Island, Subantarctic Island region of New Zealand. Burton’s 1930 concept of Pseudosuberites was that the ectosomal skeleton could vary from tangential to oblique to a palisade, all three dispositions being visible in a single specimen of what he called P. sulcatus from Campbell Island and South Georgia ( Burton 1930). However, Suberites ramosus and S. anastomosus strongly confirm to the genus Protosuberites as described above and should be transferred there: The surface of both species is described by Brøndsted (1924) as having ‘a dermal membrane (that) seems to be sustained by a layer of smaller spicules at an oblique angle, lying with the apices outwards directed, often appearing as tufts, continuing the spicule fibres’ ( S. ramosus), and ‘a dermal membrane sustained by rather compact and closely placed spicule tufts, directed more or less perpendicular to the surface; the small tylostyles compose mainly the tufts, the bigger one mainly the fibres’ ( S. anastomosus). Suberites incrustans, currently accepted as a synonym of P. sulcatus ( Van Soest et al. 2018a), is quite different from the other two species, and P. sulcatus, as it lacks a discernible ectosomal skeleton. Without direct examination of the specimen, this species is unrecognisable within Suberitidae, except for possible affinity with Prosuberites Topsent, 1893 whose species are very thinly encrusting with no ectosomal specialisation.

Pseudosuberites thurberi sp. nov. differs considerably from the shallow-water sponges described as P. sulcatus by Bergquist (1968): the tylostyles in P. thurberi sp. nov. are about 200 µm longer, the ectosome is strongly tangential (vertically disposed in Bergquist’s specimens), and they are restricted geographically to the Hikurangi Margin where they occur between depths of 900–1100 m. The new species is also differentiated from the type species, P. hyalinus and P. hyalinus var. compactus Hentschel, 1914 (both from Chile), and P. nudus from Antarctica, on the much larger size of the spicules in these three species, being well over 1000 µm long ( P. thurberi sp. nov.: max 530 µm long).

The only species that approaches P. thurberi sp. nov., at least in terms of spicule length, is P. exalbicans Topsent, 1913 from the South Atlantic Ocean, with spicules that range from 130–340 µm long ( P. thurberi sp. nov.: 291 (134–530) µm long), but it has a thickly encrusting, spreading form with a deeply wrinkled, furrowed surface. The only other New Zealand species known (with certainty) to inhabit deep New Zealand and South Atlantic waters is Cercicladia australis Rios, Kelly & Vacelet, 2011, a remarkable carnivorous sponge that has been found in only two subpolar locations on either side of the globe, on the Macquarie Ridge to the southwest of New Zealand ( 1060–1408 m) and off the Argentine coast of Patagonia ( 1145–1728 m). The authors expressed confidence in the disjunct distribution because the diagnostic spicules are unique globally and very easily recognizable. This is not the case for Pseudosuberites thurberi sp. nov. and P. albicans, which have identical megasclere types, no diagnostic microscleres and non-specialised skeletons. Pseudosuberites thurberi sp. nov. is the first accurate record of the genus in New Zealand waters and inhabits a highly specialised environment.

Pseudosuberites species are thinly to thickly encrusting with lobes and digits. The skeletons of encrusting species are usually confused, or with loosely constructed tracts of megascleres diverging from the base, spreading at the surface and extending just beyond the dermal membrane, which consists of a layer of tangentially disposed spicules. The development of lobes is usually accompanied by some degree of axial compression in the lobe. The thinner the specimen, the simpler the skeleton. Megascleres are always subtylostyles to tylostyles in two loosely differentiated categories, the larger forming the choanosomal tracts and the smaller forming the tangential ectosome. It will be important in the future to conduct a careful study of the species in this broadly defined genus, in which subgroups are currently not clear, but such a study is beyond the scope of this contribution.