Lissodendoryx (Ectyodoryx) diegoramirezensis Fernandez, Cárdenas, Bravo, Lôbo-Hajdu, Willenz & Hajdu, 2016, sp. nov.

Lissodendoryx (Ectyodoryx) diegoramirezensis sp. nov.

(Tab. 2; Figs 10–13)

Holotype. IZUA–POR 169, Isla Diego Ramírez, southern Chile (56ºS, off Cape Horn), ca. 2000 m depth, September 2002, bycatch from longline fisheries, Leg. A. Bravo. Fragment from holotype deposited under MNRJ 10884.

Comparative material. Lissodendoryx (Ectyodoryx) anacantha (Hentschel, 1914) (ZMH S 2327, fragment from the holotype).

Diagnosis. Seemingly erect Lissodendoryx (Ectyodoryx) with irregular surface, terminally microspined tylotes (223–252/7.2–9.6), acanthostyles (I. 435–602/24–31, II. 320–415/19–21, III. 220–242/14–17), arcuate isochelae with lateral alae bent towards the shaft (22–24), and sigmas (I.45 –60, II. 18.6–25).

Description. The holotype is incomplete and therefore imprecise to describe its gross morphology. It is seemingly erect, possibly stalked (Fig. 10A), ca. 2 cm high and 0.4 cm thick. Surface rather hispid, irregular, bearing short projections (Fig. 10B) and roundish openings (ca. 0.1 cm across), which are of doubtful oscular in function. A gross reticulation is apparent in the entire sponge under transmitted light in a stereo microscope (Fig. 10C). Colour in life not reported, dried holotype beige. Consistency relatively compressible and delicate, texture slightly rough.

Skeleton. Plumoreticulate with a slight degree of axial compression (Figs 11 A–B). Larger acanthostyles core main paucispicular tracts (approx. five spicules across) that are entirely echinated by another two distinct types of acanthostyles (Fig. 11 C). Some of the larger acanthostyles fan out from the main tracts, interconnecting them. Aniso-subtylotes are scattered at the surface, disposed at various angles to it (Fig. 11 D). Isochelae and sigmas are scattered in the ectosome and choanosome. There is no spongin enveloping the fibres and subectosomal lacunae are absent. Choanosomal cavities (ca. 50 µm in diameter) are uncommon, scattered and roundish.

Spicules. Megascleres (Tabs 2 –4): Aniso-subtylotes to aniso-strongyles (Figs 12 A–B), straight, microspined ends, 223– 233.5 (10.2) –252/7.2– 8.4 (1.3) –9.6. Acanthostyles I (Figs 12 C–D), straight to slightly curved, stout and somewhat fusiform; roundish base, frequently narrower than the central portion of the shaft; gradually thinning point; surface with few straight spines (up to 2.5 µm), mainly over the basal third of the spicule, 435– 536 (52.8) – 602/24– 26.6 (2.4) –31. Acanthostyles II (Figs 12 E–F), same overall morphology as the preceding category, but smaller, and bearing more spines, although the apical portion is also totally smooth, 320– 391 (29.2) –415/19– 20 (1) –21. Acanthostyles III (Fig. 12 H) straight to slightly curved; narrow roundish base; gradually thinning point; shaft completely covered with spines (moderately to abundantly), spines frequently straight and usually larger than those of the preceding categories, 220– 232.5 (10) –242/14– 15 (1.1) –17. Microscleres (Tabs 2 –4): Arcuate isochelae (Fig. 12 I), shaft gently curved, lateral alae elongated and curved towards the shaft as a claw, frontal alae simple, 22– 23.5 (1) –24. Sigmas I (Fig. 12 J), contorted, smooth, with sharp ends, 45– 55.5 (4.9) –60. Sigmas II (Fig. 12 K) same morphology as the preceding one, but smaller, 16.8– 20.8 (2.9) –25.

Ecology. Deep-water habitat (2000 m depth).

Distribution. Currently only known from the type locality.

Etymology. The species is named after its type locality.

Remarks. Lissodendoryx (Ectyodoryx) diegoramirezensis sp. nov. is distinguished from Lissodendoryx (E.)

spp. occurring in the SE Pacific, additional allied biogeographic provinces, as well as L. (E.) ballena sp. nov., L. (E.) corrugata sp. nov., and L. (E.) coloanensis sp. nov., mainly by its possession of three categories of acanthostyles, separable by their dimensions, morphology and placement in the skeletal architecture (Tab. 2; Figs 12 C–H).

Lissodendoryx (Ectyodoryx) diegoramirezensis sp. nov. has arcuate isochelae of uncommon morphology, with lateral alae of both extremities curved towards the shaft (Fig. 12 I). Precisely this isochelae morphotype can be found in L. (E.) anacantha (Hentschel, 1914: p.107, Taf. VII, Fig. 12), a species amply distributed around Antarctic and the Subantarctic Region, and in another Antarctic species, L. (Lissodendoryx) styloderma Hentschel, 1914 (Hentschel, 1914: p.101, Taf. VII, Fig. 7). This latter species can be readily distinguished from the new species by its lack of sigmas, possession of styles in the ectosome, and acanthostyles of a single category only. However, the similarity of the isochelae of the new species and those seen in L. (E.) anacantha might indicate a closer affinity between both species, as they further share similar habit and additional spicule characteristics. Nevertheless, both species are considered distinct on the basis of the presence of a third category of acanthostyles in the new species, while its acanthostyles, as a rule, are also more densely spined or with spines more widespread over the shaft. Furthermore, L. (E.) anacantha has ectosomal megascleres that can be perfect styles (Fig. 13 C–D), and the smaller category of acanthostyles may bear secondary microspines over the main spines seen at the base of the spicule (Fig. 13 E–F). These characteristics of the acanthostyles have not been spotted in the holotype of L. (E.) diegoramirezensis sp. nov. (Fig. 11 C–H). The re-examination of the type material of L. (E.) anacantha (Fig. 13) confirmed the spicule set originally reported by Hentschel (1914), and subsequently recognized by Koltun (1964). Nevertheless, new measurements obtained from the holotype (Tab. 2) revealed that acanthostyles can be much thicker than reported originally (Hentschel, 1914), matching more closely the measurements reported by Koltun (1964).

Given the data at hand, L. (E.) diegoramirezensis sp. nov. is considered well distinguished from allied forms, but it is suggested that study of additional samples of L. (E.) anacantha, such as those registered (but not described) by Burton (1932) from South Georgia, is important to verify if these two taxa are distinct, as hypothesised here.