Haliclona ( Soestella) battershilli sp. nov.

( Figs 1–4; Table 1, 2)

Haliclona n. sp. cf. kaikoura, Stoffers et al. 1999: 66, 247.

Material examined. Calypso hydrothermal vent field ( Southern vent field, unless noted otherwise), southwest of White Island, Bay of Plenty: Holotype — NIWA 32128, University of Kiel Stn SO192-2/2, 37.688° S, 177.122° E, 189 m, collected by ROV, 24 Apr 2007. Paratypes — NIWA 52850, IFM GEOMAR Stn SO135/103, 37.695° S, 177.101° E, 179–181 m, collected by rock dredge, 10 Oct 1998; NIWA 52895, IFM GEOMAR Stn SO135/102, 37.698° S, 177.099° E, 177–185 m, collected by rock dredge, 9 Oct 1998; NIWA 52921, IFM GEOMAR Stn SO135/110, 37.688° S, 177.122° E, 189–192 m, collected by rock dredge, 11 Oct 1998; NIWA 52872, IFM GEOMAR Stn SO135/108, 37.687° S, 177.121° E, 193 m, collected by submersible, 10 Oct 1998.

Other material. Calypso hydrothermal vent field ( Northern vent field): NIWA 99603, IFM GEOMAR Stn SO 135/117, 37.614° S, 177.100° E, 162 m, collected by rock dredge, 11 Oct 1998.

Type location & distribution. Calypso hydrothermal vent field, Bay of Plenty, New Zealand, 162–193 m.

Description. Sponge forms a tangled mass emanating from what appears to be either a single point of attachment to hard substrate, or an encrustation ( Figs 2, 3A; see also Fig. 5A). Immature sponges form a solitary, short stubby to elongate irregular finger, often branched into two or three fingers, length about 10–20 cm long, 4– 10 mm thick. As the sponge grows, the basal branch splits, branches anastomosing to form a large tangled mass, up to about 40 cm long and 30 cm wide. Oscules, 2–5 mm diameter, are raised on low mounds on older, basal branches, rendering the surface knobbed, almost ribbed in appearance (see Fig. 5A). The surface of younger sections is smooth and oscules are flush with the overall surface, slightly concave in preservative. Texture in life slightly compressible, younger sections more fragile and compressible, older, basal sections firmer slightly elastic and reasonably flexible. Brittle on tearing. Surface texture slightly fuzzy, microscopically hispid. Colour in life light cream; in preservative pale tan.

Skeleton. Skeleton of the deep choanosome consists of rare, uni- to multispicular primary lines ( Fig. 3C, red arrows), joined in places by single spicules ( Fig. 3C, white arrows), with evidence of some minor subisodictyal reticulation. The majority of the choanosome has little order, rather, consisting of abundant, scattered spicules, surrounding subdermal spaces of variable size. The ectosome is highly irregular with spicules protruding obliquely or with no obvious orientation, creating the slightly velvety surface texture. Spongin is not visible but most likely limited to the nodes of connection between oxeas.

Spicules. Megascleres ( Table 1; Fig. 4) thick oxeas, fusiform, slightly curved to abruptly, slightly centrally bent, 203 (159–231) × 11 (5–15) µm.

Substrate, depth range and ecology. Attached to volcaniclastic rock covered in sediment and associated with high densities of an undescribed orange anemone, and other sponges, 177– 193 m.

Etymology. Named for Professor Christopher N. Battershill, University of Waikato, who first identified this species as Haliclona n. sp. cf. kaikoura Bergquist & Warne, 1980 and who noted the potential significance of the association between the sponges and active hydrothermal venting at the Calypso hydrothermal vent field ( Stoffers et al. 1999).

Remarks. In general terms, the skeletal architecture of Haliclona battershilli sp. nov. conforms to something between that of Haliclona ( Haliclona) species sensu de Weerdt (2000, 2002), which have a very regular ladder-like reticulation of uni- to pausispicular primary lines connected by unispicular secondary lines, and Haliclona ( Soestella) de Weerdt, 2000 species, which have a subanisotropic choanosomal skeleton of ill-defined pausispicular primary lines, irregularly connected by pausispicular secondary lines, with a slight tendency to form rounded meshes (de Weerdt 2000, 2002). The choanosome of Haliclona battershilli sp. nov. has only occasional unispicular primary lines visible, joined occasionally by single spicules ( Fig. 3C) [as in Haliclona ( Haliclona)], interspersed with numerous interstitial oxeas strewn around a reasonably cavernous choanosome. The skeleton approaches Haliclona ( Soestella) in this way but does not form strictly rounded meshes. In the face of these difficulties, we consider the skeleton architecture to be closer to that of Haliclona ( Soestella) than to that of Haliclona ( Haliclona) species and have assigned the new species to the subgenus Haliclona ( Soestella) accordingly.

Stoffers et al. (1999: 66) describe, “a large very dense anemone assemblage with white finger sponge ( Haliclona n. sp.) together with bacterial mats” as, “a striking assemblage,” that was listed as specific to the venting areas ( Stoffers et al. 1999: 72; Species/geological associations: 1–white bacterial mat, brown anemone and white finger sponge/Rock outcrops, bubbles, high temperatures). The ‘white finger sponge’ is named in Stoffers et al. (1999: Appendix 6-3) as Haliclona n. sp. cf. kaikoura Bergquist & Warne, 1980, which forms branches that arise from a flattened base, but this species has a conspicuous punctate, reticulate surface and much smaller oxeas [ H. kaikoura: 130 (124–149 µm); H. ( S.) battershilli: 203 (159–231 µm)]. Haliclona kaikoura has only been recorded from the east coast of the South Island.

Port Ross and Carnley Barrel-shaped, tubular to encrusting, Irregular meshes; primary Oxeas, slightly bent in the middle, Bergquist & Warne (1980) report a “densely Harbour, Auckland Is, up to 40 mm high, 15 mm wide, fibres more distinct than occasionally centrotylote spiculose skeleton, making their placement in R. Subantarctic New encrusting up to 55 mm wide, 1–3 secondary fibres; main 130–135 × 6.6–8 µm Auckland Is cinerea doubtful.” The species was first and Zealand, intertidal mm thick; oscules 1–1.5 mm fibres often 3 spicules 125–160 µm Stewart Is (Bergquist & subsequently described from the North Atlantic and (Brøndsted 1924); diameter, margins raised; surface wide and project through Warne 1980) Mediterranean Sea. The New Zealand distribution is Slipper Is, low water; finely hispid; texture softly elastic; ectosome; spongin scarce; inaccurate (Van Soest 2017d) and the Slipper Island Port Pegasus, Stewart Is, colour in life, yellowish-grey to megascleres oxeas (Coromandel Peninsula?) record, disparate with the intertidal (Brøndsted brownish (Brøndsted 1924) (Brøndsted 1924) generally Subantarctic New Zealand distribution 1924)

Haliclona ( Reniera) clathrata (Dendy, 1895) sensu Brøndsted (1924: 125; 1924: 453), Bergquist (1961: 35; 1961b: 170) and Bergquist & Warne (1980: 14)

Perseverance Harbour, Irregularly massive, raised, conical, Very regular; primary Oxeas, slightly curved in the middle, The species was first described from Port Philip Campbell Is, funnel-shaped oscules, 0.6 mm fibres, 2 (1–3) spicules occasionally centrotylote Heads, Victoria, South Australia (Dendy 1895) and Subantarctic New diameter; sponge 12 mm diameter; wide, radiate 83 × 5 µm Port Philip (holotype) subsequently from Port Jackson, New South Wales Zealand, intertidal surface finely hispid; texture soft, perpendicular to the 107 × 6 µm (holotype remeasured) (Brøndsted 1924). Bergquist & Warne (1980) (Brøndsted 1924); elastic, colour in life, pale grey surface; connected by 85–115 × 5 µm Campbell Is considered the species to be well characterised. Queen Charlotte Sound, (Brøndsted 1924). Massive single spicules (Brøndsted 90–105 × 6 µm QC Sound However, because of the purportedly huge latitudinal 1–3 m (Brøndsted encrusting sponge with broad-based 1924). Dense, more or less 136 (120–141) × 6 (5–6) µm Chatham range of this species and the considerable range in 1924); 6–20 m conical elevated oscules; 6 cm long, irregular unispicular Is the length of the oxeas, we consider this species to (Bergquist & Warne 4 cm wide, thickness 2.8 cm; surface reticulation. Distinct 108 (92–124) × 5 (4–7) µm Ladies either a possible adventive from South Australia, or (1980); Waitemata smooth; texture soft, friable, large multispicular primary Bay (Bergquist & Warne 1980) more likely, a mix of species that might be resolved Harbour; Wellington; specimens are elastic; colour in life, fibres are visible in places, by examination of the original and fresh material in Kaikoura; Chatham Is; yellow-brown to mauve (Bergquist projecting from the comparison with the South Australian holotype Akaroa Peninsula; & Warne 1980) ectosome (Bergquist &

Dunedin (Bergquist & Warne 1980)

Warne 1980)

……continued on the next page

……continued on the next page

……continued on the next page

Perseverance Harbour, Almost barrel-shaped, 135 mm long, Primary fibres, 2–6 Oxeas, slightly bent in the middle with Haliclona laxa was first described from Iceland by Campbell Island, 37 m; 7–8 mm thick; oscule at end of spicules thick, radiate a sharp curve Lundbeck (1902) and the name attributed to Port Ross, 18 m, barrel, 1.5 mm diameter, surface towards the surface, 182 × 10 µm Perseverance Harbour specimens in New Zealand by Brøndsted (1924). Auckland Island, finely hispid; texture soft, elastic; connected irregularly by 190–200 × 7–8 µm Colville Channel Haliclona laxa sensu strictu is a synonym of the Subantarctic New colour in life, very light yellowish single spicules. Overall northern European species H. ( Rhizoniera) Zealand (Brøndsted grey (Brøndsted 1924). Colville skeleton very dense with rosea (Bowerbank, 1866) (Van Soest 2009) and it 1924) Channel specimen encrusting 1–3 numerous interstitial highly unlikely that the species name is appropriate mm thick on a shell (Brøndsted spicules (Brøndsted 1924). for our New Zealand specimens, given the bipolar Colville Channel, 1924) In the northern specimen, nature of the purported distribution. Furthermore, the Hauraki Gulf, 64 m the primary fibres are not distribution within New Zealand is disparate, with (Brøndsted 1924) visible due to the locations at either end of the country (Subantarctic encrusting habit New Zealand vs Auckland region) and the skeleton (Brøndsted 1924) in both specimens differ, with the Colville Channel specimen having considerably larger oxeas. The northern and subantarctic Brøndsted specimens are highly likely to be distinct species and both are possibly new endemics, the identity of which can only be determined with a study of fresh material.

……continued on the next page ……continued on the next page

……continued on the next page

In terms of spicule dimensions, most New Zealand Haliclona species have oxeas that range in length from about 100–150 µm ( Table 2); H. sabulosa Bergquist & Warne, 1980, has the smallest oxeas, at 81 (68–95) × 5 µm and H. maxima Bergquist & Warne, 1980 has the largest, at 293 (274–317) µm long. With oxeas that range in length from 159–231 µm, H. ( S.) battershilli sp. nov. has the second largest spicules recorded for a species of Haliclona in New Zealand.

The three species most closely comparable to H. ( S.) battershilli sp. nov., in terms of oxea dimensions, are H. fragilis ( Bergquist & Warne, 1980), H. ‘laxa’ ( Lundbeck, 1902) sensu Brøndsted (1924) [ H. laxa ( Lundbeck, 1902) is now accepted as Haliclona ( Rhizoniera) rosea ( Bowerbank, 1866) in the World Porifera Database (van Soest et al. 2018d)], and H. ( S.) implexa (Schmidt, 1868) sensu Brøndsted (1924: 122) and Bergquist & Warne (1980: 17). Haliclona fragilis has oxeas 175 (131–197) µm long, and has a loose, unispicular reticulation with visible spongin, and has only been recorded from the lower intertidal ( Auckland west coast) and shallow subtidal (South Island east coast). Haliclona ( R.) rosea from Colville Channel in the Hauraki Gulf, is probably the closest species to H. ( S.) battershilli sp. nov., with oxeas 190–200 µm long, and recorded from 64 m depth. However, H. ( R.) rosea was described as tubular, about 14 cm long, with a wall about 8 mm thick. The skeleton is also similar, described as very dense with numerous interstitial spicules and composed primarily of primary fibres, 2–6 spicules wide. The primary fibres in H. ( S.) battershilli sp. nov. are rare and unispicular. Haliclona ( S.) implexa was recorded from shallow waters in the Subantarctic region of New Zealand, but Brøndsted’s description does not conform to that of Soestella sensu stricto ( Table 2), and the species is now reserved for Northern Atlantic and Mediterranean sponges; the New Zealand distribution is considered to be inaccurate ( Van Soest 2017f).

Despite the general difficulty of differentiating the many species of Haliclona in New Zealand waters, we regard H. ( S.) battershilli sp. nov. as unique, due to: 1) the restricted distribution to the Calypso hydrothermal vent fields; 2) the depth of this habitat ( 180–190 m), the deepest record for any species of Haliclona in New Zealand waters thus far; 3) the considerable size of the oxeas (159–231 µm) compared to most other species; and 4) the branching, ramose, attenuated habit, not found in any other New Zealand species.