Carnivorous sponges (Porifera, Cladorhizidae) from the deep South Pacific (New Caledonia) with the description of three new species of the genus Abyssocladia and remarks on genus Cercicladia Author Vacelet, Jean text Zootaxa 2020 2020-04-24 4767 2 257 276 journal article 22581 10.11646/zootaxa.4767.2.3 cab7562c-80d6-4c50-be4c-1818403c0c4b 1175-5334 3770819 urn:lsid:zoobank.org:pub:6A6A15AA-9B6D-4168-8857-29E183D965B9 Cercicladia australis Rios, Kelly & Vacelet, 2011 Material examined . KANACONO Expedition, New Caledonia ( MEOW ), Station beam trawl CP4753, 25/08/2016 , 23°18’S 167°57’E , 758– 952 m . One specimen. Description . A single specimen of an erect sponge, morphologically similar to the specimens of the original description, but incomplete and broken into three fragments. The sponge is composed of a flattened axis, 130 mm long and 4–5 mm in width, 1.5 mm thick, broken at the base and devoid of an attachment base. The axis is smooth for 50 mm near the base, then bears biserially arranged short, conical buds, 1–3 mm long by 1–2 mm in diameter at the base, that are often broken at their apex and devoid of filaments. The texture is firm, but breaks easily. The color is brown in ethanol. There is no visible aperture or canal system. Skeleton. The stem and buds have an axis of longitudinally arranged styles. The styles of the axis of the buds are similar to those of the main axis, only slightly shorter and not exceeding 1550 x 32 µm , whereas they measure up to 2150 x 38 µm in the axis. The microscleres are dispersed without apparent order in the poorly preserved living tissue. Cercichelae and sigmancistras are rare in the axis, but extremely abundant in the buds. Microxeas are moderately abundant everywhere. Toxas are rare in the buds, but more abundant in the axis. Spicules ( Fig 9 ). Styles ( Fig 9B ): straight, slightly fusiform, with a short rather blunt point and a short restriction near the head, 1125–2150 µm x 30–38 µm , slightly smaller in the axis of the buds where they do not exceed 1550 µm x 32 µm . Microxeas ( Fig 9C ): generally smoothly curved in the middle but often straight, with long acerate ends, 120– 190 µm x 3.5–5 µm . Toxas ( Fig 9D ): more or less angulose, but often straight, 45–55 µm . These microscleres do not have a true shape of toxa, but may also be considered as microxeas. Some intermediaries with the larger microxeas, less angulose, are present. Sigmancistras ( Fig 9E ): extremely abundant in the buds, C- shaped, thick with short recurved, acerate tips in the same plane, without notches, 27–40 µm x 3.5–4 µm . Cercichelae ( Fig 9F ): very abundant in the buds where they are dispersed without order, very similar to those of the original description. A suture line between the two teeth is sometimes visible, 72–100 µm x 12–20 µm . Distribution and Ecology. Collected by trawl, 758–952 m depth. Remarks. This sponge has the same morphology as in the original description of C. australis . The only difference is the absence of a fleshy pedicle at the base, which is evidently incomplete. The thin, fragile filaments present at the end of the lateral buds in some paratypes of the species and visible in in situ pictures of C. australis are also missing, as is the case for most specimens of the original description, which is also likely due to the state of preservation. The spicule complement of this specimen from New Caledonia also shows strong similarities with that of C. australis . It is well characterized by the presence of a special type of spicule, the cercichelae, by the absence of true chelae, and by the presence of sigmancistras, microxeas and small microxeas which may be considered as toxas as suggested by Rios et al . (2011) . The cercichelae are here exactly similar to those of C. australis . Acanthosubtylostyles are absent, but this is very likely due to the absence of a pedicle in this incomplete specimen. The sponge is thus evidently a member of the very original genus Cercicladia . However, the question of its specific status needs to be examined. This question initially arose for the first description of the genus, in which the specimens had surprisingly very different geographic origins, some being from the Macquarie Ridge of New Zealand , in the Pacific, and others from the Argentine coast of Patagonia, in the South Atlantic. It was concluded at that time that the few differences in the spicule complement did not justify the distinction of two species. The new specimen from New Caledonia , although relatively near New Zealand , allows to shed new light on the problem. Our specimen shows some differences with C. australis , in both the New Zealand and the Patagonia specimens. Lacking are acanthosubtylostyles, spicules localized in the pedicle which is rarely preserved, but this was general in the New Zealand specimens and also in most of those from Patagonia. The mycalostyles appear smaller in the specimen from New Caledonia , especially when compared with the Patagonian one: 2150 x 38 µm in the New Caledonian versus 2433 x 84 µm in the Patagonian specimen. It appears however that the very large thickness of 84 µm stated in the original description of specimens from Patagonia is a typing error. After verifying the slides the thickness of these mycalostyles was found to not exceed 54 µm . The microxeas are slightly curved in the New Caledonia , straight in the type specimens. The sigmancistras are devoid of notches, which are faint and not always present in the types . There are a few other slight differences in spicule size. The difference in mycalostyles thickness appears not significant after correction of the typing error to 54 µm , rather than 84 µm , as initially stated for specimen 4096 from Patagonia. These differences do not currently justify the distinction of a new species, but it should be noted that the specimen from New Caledonia appears more similar to those from New Zealand than to those from Patagonia. FIGURE 9. Cercicladia australis . A—Specimen CP-4753. B—Style. C—Microxea. D—Toxa. E—Sigmancistras. F—Cer-cichelae. G—General view of the spicules. General remarks The discovery of these three new species of Cladorhizidae shows once again that cladorhizid sponges diversity in the deep sea, and particularly in the deep Pacific, is far higher than was thought fifty years ago. These three species are only a portion of the Cladorhizidae collected south of New Caledonia during the KANACONO and KANADEEP campaigns. The remaining Cladorhizidae collected include other genera under study, with several probable new species. The recent interest in carnivorous sponges and in the deep-sea in general explains this increase in knowledge concerning these sponges. According to several papers in press or in preparation, new advances in the inventory of deep-sea cladorhizids are to be expected soon, especially in the deep Pacific. The three new Abyssocladia described here also increase knowledge concerning the diversity of spicules of car- nivorous sponges, especially for cheloid microscleres. Macrosclere spicules in general have a skeletal or a protective role. The function of microscleres, however, is less clear and very largely unknown. The role of chelae and most other microscleres in filtering, non-carnivorous sponges is far from well understood—if we forget the invaluable help of these spicules to the sponge taxonomists for identification–. Yet in spite of lacking any function—or at least any or known function—the cheloid spicules have nevertheless greatly diversified in filtering poecilosclerids. Very remarkably, in carnivorous sponges these spicules have acquired a precise function in the capture of prey. These sponges build a regular arrangement of chelae on filaments or inflated spheres, with their hook-like teeth and alae outwardly exposed, acting as a Velcro-like cover for ensnaring the setae or appendages of their prey. This mode of capture of macro-prey has been evidenced in several species ( Lee et al . 2012 ; Vacelet & Duport 2004) and this is a remarkable example of exaptation (“ features that now enhance fitness but were not built by natural selection for their current role” ( Gould et Vrba, 1982 )), i.e. the process of acquiring functions to which they were not originally adapted. It must be expected that chelae, already very diversified in filtering sponges, markedly increase their diversification in Cladorhizidae in which they develop a new function. There are many known examples, as underlined by Vargas et al . 2012 , and further cases are added by the new species described here. However, a problem with A. mucronata sp. nov. is that the shape of the chelae, with its intermingling appendages, appears poorly adapted to the trapping of prey. At first glance, it thus seems that this diversification in shape would not facilitate feeding.A similar problem occurs in A. microstrongylata sp. nov. , in which the axis and appendages bear a dense coating of microstrongyles, which appears poorly designed for catching the appendages of a prey. Similarly, several Abyssocladia species, including the type species A. bruuni , have abyssochelae with appendages which are nearly in contact. The cercichelae of Cercicladia also appear poorly designed, contrary to many chelae or the placochelae of Euchelipluma . This problem was raised by Watling (2007) for Cladorhiza corona Lehnert, Watling & Stone, 2005 , in which the chelae were not arranged in a manner suitable for prey trapping, but which nevertheless contained prey in various stages of digestion. Watling hypothesized that the prey, mostly copepods, might be trapped by means of a sticky substance. In the present study, no prey in the process of being digested were observed in the available specimens of these sponges, which may be explained by the rarity of prey in such deepsea environments. Indeed, in the same collection, prey were also not observed in A. kanaconi , which has the usual Velcro-like lining. In A. mucronata , the poorly preserved living tissue contained numerous fragments of foreign debris, which may bring to mind the hypothesis of Michael Sars ( Sars, 1872 ) in the first description of a cladorhizid, Cladorhiza abyssicola Sars, 1872 , that foreign particle may make up at least part of its nourishment. Carnivory is thought to be general in Cladorhizidae . This mode of life has been well documented in the few species that are easily accessible for observation or experimentation, and prey in the process of being digested has been observed in several deep-sea species. But species such as A. mucronata sp. nov. or A. microstrongylata sp. nov. show that more remains to be discovered on the mechanism of feeding in deep sea Cladorhizidae .