Introduction

Carnivorous sponges, family Cladorhizidae, are unique in the phylum Porifera in possessing a combination of skeletal and organic tissue properties allowing them to catch small prey animals such as copepods and cladocerans (Vacelet and Boury-Esnault 1995). With the exception of the genus Chondrocladia Thomson (1873), the sponges of this family lack an aquiferous system. The overwhelming majority of carnivorous sponges occur in deep sea soft bottom habitats, and their deviating feeding strategy and accompanying physical features are explained as an adaptation to paucity of filter feeding biomass in oligotrophic deep sea habitats. The carnivorous feeding strategy was first detected in Asbestopluma hypogea Vacelet and Boury-Esnault (1996), which in contrast to most other cladorhizids lives in a shallow-water rocky habitat (caves near Marseille). Nevertheless, this habitat is considered similarly oligotroph and considered a displaced deep sea environment allowing the colonization of deep sea specialist sponges. Recently, SCUBA exploration of Antarctic rocky habitats near McMurdo Sound, yielded approximately 45 species of sponges (B. Baker, personsal communication). Among these shallow-water sponges, surprisingly, a species of Asbestopluma was found growing in a richly populated semi-open cave habitat at 30 m depth. The genus Asbestopluma counts 31 species (Van Soest et al. 2008), including two species of the subgenus Helophloeina Topsent (1929). Comparison with descriptions of Asbestopluma species from Antarctic waters and elsewhere led to the conclusion that the species is new to science. It is the purpose of this study to describe and name this species. The presumed oligotrophy cannot be confirmed for the habitat of this species which calls for additional considerations concerning the carnivorous feeding strategy.

Materials and methods

Sponge material studied here is deposited in the collections of the Zoological Museum of the University of Amsterdam (ZMA). For microscopic examination, a fragment of the sponge was dissolved in concentrated nitric acid and the residue was rinsed several times in distilled water. The resultant spicule suspension was mounted on SEM stubs for examination under a JEOL Scanning Electron Microscope and on several glass slides for light microscopic examination. In addition, thick sections were mounted on glass slides for the study of the skeleton structure. Digital images of the spicules were assembled on a black background and aligned and cleaned up using Adobe Photoshop CS3. Thick sections were photographed using a Nikon photomicroscope. Spicule sizes (smallest–average–largest) were determined from 25 light microscopy measurements for each spicule category.

Systematic Descriptions

Class Demospongiae

Order Poecilosclerida

Suborder Mycalina

Family Cladorhizidae

Genus Asbestopluma Topsent, 1901

Subgenus Asbestopluma Topsent, 1901 sensu Vacelet, 2006

Asbestopluma (Asbestopluma) vaceleti n.sp.

Figs. 1, 2, 3

Fig. 1
figure 1

Asbestopluma (Asbestopluma) vaceleti n.sp. a Sponge in situ at Turtle Rock, McMurdo Sound, 30 m (photo, B. Baker), b holotype ZMA Por. 13264, habit, c ending of one of the holotype branches showing skeletal 'filaments' (scale bar 500 μm), d detail of surface showing trapped cladoceran (scale bar 100 μm)

Full size image
Fig. 2
figure 2

Asbestopluma (Asbestopluma) vaceleti n.sp. Detail of branch ending and swollen apices of filaments of in situ specimen (photo, Shawn Harper)

Full size image
Fig. 3
figure 3

Asbestopluma (Asbestopluma) vaceleti n.sp. a Style of the central axis, and details of apices, b style of the surface bundles and detail of tyled head, c forceps with straight and trailing legs, d detail of swollen end of forceps leg, e anisochelae 1 from side and back, and detail of serrated lower ala, f sigma-like anisochela 2

Full size image

Synonymy

Asbestopluma aff. lycopodina (sic); Brueggeman 1998: 9.

Material examined

Holotype ZMA Por. 13264, Antarctica, McMurdo Sound, Turtle Rock, 30 m, coll. B. Baker, nr. ASC-VI-97-093, SCUBA, June 1997.

For comparison

ZMA Por. 19989, identified as Asbestopluma lycopodium, SE Rockall Bank, 573 m, coll. R.W.M. van Soest during BIOSYS 2005, from boxcore sample BX153, 11 July 2005.

Description

Tangled white stalks hanging down from rocky overhangs (Fig. 1a), with regular fringing surface projections (‘filaments’) arranged in bottle brush fashion. Size of individual branches of in situ specimens up to approx. 15 cm × 3–5 mm; in the preserved holotype branches (Fig. 1b) are approx. 50 × 3 mm. Filaments are 1–1.2 mm in length and lie at distances of 0.3–0.5 mm. At the top stalks are slightly swollen, some more so than others, and filaments on and near the top show swollen apices. Basal branching is indistinctly visible (Fig. 1a), but further along the branches there is only occasional division and predominantly branches are undivided. Overall impression is single hanging stalks with little variation in length or diameter. Colour a bright shining white.

Although the collected material did not contain any basal attachments, the in-situ photograph (Fig. 1a; small sponge in the middle part) shows rather indistinctly that the holdfast is small and is not an extensive root system. Skeleton (Figs 1c, d): long styles are arranged parallely in the centre of the stalks surrounded by tissue containing forcipes and chelae. At the surface, there is a dense lining of chelae. From the central axis at regular intervals, bundles of styles are sticking out at right angles forming the axis of the filaments (Fig. 1d). At the top of the stalk, the central axis fans out slightly and forms a dense bush of short spicule bundles sticking out beyond the surface (Fig. 1c).

The swollen endings of the terminal filaments (Fig. 2) have become shrunken in the preserved material to the extent that they are no longer clearly visible on the filaments. However, whereas most surfaces have a lining of anisochelae, we observed a slight amassing of anisochelae in distal parts of many of the filaments near the top of a branch. Conceivably, the swollen endings of the filaments were lined with anisochelae and, when they collapsed after the sponge was taken out of the water, they caused this increased amount of anisochelae near the end of the filaments.

Spicules

Styles in two size classes, chelae in two shapes, forcipes.

Styles of the central axis (Fig. 3a), with narrow rounded heads and slightly swollen along the shaft (‘fusiform styles’), length × width 1,210–1,408.2–1,645 × 11–18.3–30 μm.

Styles of the surface bundles (Fig. 3b), similar in shape but frequently with a faint tyle, 271–561.6–822 × 4–11.0–18 μm.

Forceps (Fig. 3c), arranged in rosette-like groups, legs often tangled or trailing to either side, legs provided with a terminal knob (Fig. 3d), 33–45.6–69 μm.

Anisochelae in two shapes that were initially assumed to be growth stages of a single microsclere category. However, no real intermediates of the two types were found and their frequency of occurrence, though different, was high for both. Type 1 palmate anisochelae of the usual Asbestopluma type (Fig. 3e), with lateral upper alae entrely fused with the shaft, and with the lower alae serrated, of closely uniform size, 13–14.9–16 μm. Type 2 sigma-like palmate anisochelae (Fig. 3f) with reduced upper alae and peculiar sharply chiseled lower alae, of closely similar size and shape, slightly smaller than the normal chelae, 12–12.8–14.5 μm.

Ecology

Occurring in semi-open overhangs among densely crowded hard substrate communities at shallow depth (30 m). This is one of the carnivorous sponges sequestering small crustaceans (see Fig. 1d) from the ambient water by the sticky fringed surface. There are two photos of live individuals on the Scripps Ross Island and McMurdo Sound website (Brueggeman 1998), which are here reproduced in Fig. 1a and Fig.2.

Distribution

So far only known from the type locality in McMurdo Sound, Antarctica.

Etymology

Named to acknowledge Jean Vacelet for his pioneering work on carnivorous sponges.

Remarks

No other Antarctic Asbestopluma species has the combination of characters described above. The shape, but especially the possession of forceps microscleres, is so far unique for this region. Forcipes supposedly are associated with sperm productions (Vacelet and Boury-Esnault 1996), so it could be speculated that—although unlikely—they may have been overlooked in previously reported Antarctic Asbestopluma species.

Table 1 lists all the reported Asbestopluma species from the Antarctic region. A. callitrix Hentschel (1914) is the only species that appears overall close. The habit is club-shaped with a smooth stalk but the ovoid ‘body’ appears somewhat similar to our new species although relatively much more swollen. The megascleres are up to 4 mm long, much longer than those of A. (A.) vaceleti n.sp. and the chelae are only 11–12.5 μm. No forceps are reported and depth occurrence is 350–385 m. A. calyx Hentschel (1914) is a stalked cup and differs in having smaller styles and larger anisochelae. Likewise no forceps are mentioned and depth occurrence is bathyal. A. belgicae (Topsent 1901) is similar in habit to A. (A.) vaceleti n.sp. forming long stalks with filaments along the entire stalk, but filaments are distinctly longer (13 mm) and more entangled than those of A. vaceleti n.sp. Spiculation differs sharply by the possession of rugose microtylostyles and sigmas and depth occurrence is bathyal. A. obae Koltun (1966) is in most aspects similar to A. belgicae, but it apparently lacks filaments (fragmentary specimen). A further unnamed specimen reported from abyssal depth by Koltun (1966) has indistinct habit and clearly larger megascleres than A. (A.) vaceleti n.sp.

Table 1 Spicule data of Antarctic Asbestopluma specimens
Full size table

Originally, this species was provisionally identified as A. lycopodium Levinsen (1887) because of overall similarity with this Arctic–Boreal deep sea species. Probably as a result of a misspelling, this identification was changed to Asbestopluma aff. lycopodina Lundbeck (1905) in Brueggeman (1998), an incorrect name because the subgenus Lycopodina was erected by Lundbeck for a group of similar species including A. lycopodium, but there is no species named lycopodina. We were able to compare the present material with a small fragmentary specimen, ZMA Por. 19989, identified as A. lycopodium from Rockall Bank deep sea coral reefs, W of Ireland (Van Soest et al. 2007). The two are similar in habit and megasclere spiculation (see Table 2), but A. lycopodium lacks a second shape category of anisochelae and the microsclere sizes were clearly different (smaller forcipes and chelae). The latter difference may not be significant, as microsclere sizes for this species provided in the literature are different again from both the Antarctic and the Rockall Bank specimens. In fact, A. lycopodium is probably not the most similar Asbestopluma to our new species. The shallow water Mediterranean A. hypogea Vacelet and Boury-Esnault (1996) conforms to our new species in almost all aspects, but it has a smooth stalk and ovoid body, unlike A. (A.) vaceleti n.sp., smaller megascleres, smaller forcipes not arranged in rosettes, and the second category of reduced anisochelae is lacking.

Table 2 Spicule data of Asbestopluma species possessing forceps microscleres
Full size table

From Table 2, it can be concluded that forceps microscleres occur so far almost exclusively in the northern hemisphere. An exception is A. hadalis Lévi (1964) from the Kermadec Trench. This differs from our new species in shape (swollen body midway in the stalk) and by having much larger long styles (up to 3,200 μm). Furthermore, Janussen and Plotkin (personal communication, not mentioned in Table 2) collected a possibly new species of Asbestopluma from the deep Weddell Sea (c. 4400 m depth), which has unusually thick forceps and anisochelae clearly different in shape from those of A. (A.) vaceleti n.sp.

Discussion

The shallow-water occurrence of our new species parallels that of the Mediterranean A. hypogea, but there is one clear difference: A. hypogea occurs in a ‘displaced deep sea’ environment, comparable in trophic aspects to a true deep sea environment, whereas A. (A.) vaceleti n.sp. was found in a densely populated habitat with all the characteristics of high nutrient content (see Fig. 1a). Nevertheless, from indirect indications, it is likely that A. (A.) vaceleti n.sp. still maintains its carnivorous feeding strategy of being sticky and absorbing small invertebrates (see cladoceran fixed to the surface of the sponge in Fig. 1d). If this is corroborated by more direct observations, then we must assume that the carnivorous life strategy is firmly established in Asbestopluma, as it is maintained in A. (A.) vaceleti n.sp. even though a suspension feeding strategy would be more warranted in its habitat. The carnivorous feeding strategy occurring in shallow-water habitats represents a form of niche partitioning with a distinct advantage towards others members of the sessile community.