C. pusillum sp. n., here designated.

Colonies form encrusting sheets or are lobate. Polyps monomorphic and retractile. Polyps with point spindles showing an arrangement in chevrons, a kind of collaret can be present, formed by the lowest point sclerites lying horizontally. Coenenchymal sclerites are wide spindles and ovals with simple and complex tubercles. The polyparium additionally can have clubs in the surface layer, which are derived from the spindles. When preserved, colonies are white or coloured; sclerites colourless or coloured. Azooxanthellate.

From the latin complexus, a complex, an aggregate of parts, referring to the complex tubercles common on the coenenchymal sclerites of this genus.

The following West African species hitherto placed in Alcyonium are refered to the new genus: A. caparti, A. globosum, A. gruveli, A. laxum, A. miniatum, A. monodi, A. patulum, A. pobeguini, and A. strictum; all were described by Tixier-Durivault (1955).

Holotype: Congo, “Pointe Noire”, Banc Mullet, 31.I. 2013, depth -10 m, 1 colony, (RMNH Coel. 41604); paratype: same data as holotype (RMNH Coel. 41605).

The holotype is a colony consisting of two lobes, 2 × 1.3 cm in diameter and 1 cm thick, encrusting rock (Fig. 2A). The polyps are completely withdrawn into the coenenchyme and calyces are not present.

Figure 2. 

A, B Complexum pusillum sp. n. A holotype (RMNH Coel. 41604) B paratype (RMNH Coel. 41605). Scale bar 10 mm.

The anthocodiae have a collaret composed of 2–3 rows of spindles. These spindles are up to 0.15 mm long, slightly bent, and have simple tubercles (Fig. 3B). The points have spindles similar to those of the collaret, 4–5 pairs per point. They are also up to 0.15 mm long, have simple tubercles and a slightly spiny distal end (Fig. 3A). The tentacles contain no sclerites.

Figure 3. 

Complexum pusillum sp. n., holotype (RMNH Coel. 41604). A point spindles; B collaret spindles C–D clubs of surface layer E spindles of interior.

The surface layer of the top of the colony has clubs, up to 0.35 mm long, with complex tubercles and spiny heads (Fig. 3C–D). The interior has straight and bent spindles, up to 0.60 mm long, with simple or complex tubercles (Fig. 3E).

The base of the colony has spindles and ovals, up to 0.65 mm long, with simple or complex tubercles (Fig. 4).

Figure 4. 

Complexum pusillum sp. n., holotype (RMNH Coel. 41604). A–B spindles of base of colony.

The preserved holotype is white, and all sclerites are colourless.

From the Latin, pusil, tiny, referring to the small size of the colonies.

The paratype is a single lobe-like colony (Fig. 1B). The sclerites are similar to those of the holotype.

In the studied area, coastal waters show a general high turbidity due to the input of sediment and detrital and humic materials by the Congo River in the south and the Kouilou River in the north, as well as a high primary productivity in the ocean. As a consequence muddy bottoms dominate on the continental shelf (Giresse 1980). Nevertheless, shallow, cretacean, calcareous banks emerge among them, at a depth of between 5 and 20 m (Giresse and Kouyoumontzakis 1973). These rocky banks constitute large, thin slabs (no more 1.5 m high) sometimes exposed to high hydrodynamics due to the swell, tides and the occurrence of complex exchanges of water bodies up to 80 m deep (Moroschkin et al. 1970; Piton 1988). Alcyonium pusillum sp. n. colonizes these shallow rocky banks, mainly in cryptic positions (under overhangs and in large holes) forming small white patches (< 1 m²) easily distinguishable by SCUBA divers (Fig. 5A, B). On these hard bottoms, the associated fauna is mainly composed of gorgonians (Eunicella, Leptogorgia and Muriceopsis species), stony corals (Polycyathus cf. senegalensis Chevalier, 1966 in the same cryptic positions as A. pusillum sp. n.) and sponges (Fig. 5C).

Figure 5. 

Complexum pusillum sp. n., Banc du Conflit, depth 10 m. A General view of a small patch of colonies on rocky bottom B Magnified view of some colonies C small piece of rock colonized by the new species, sponges (green arrow) and small stony corals (Polycyathus cf. senegalensis – red arrow).

The two encrusting Alcyonium species previously described from Congo, A. globosum Tixier-Durivault, 1955, and A. laxum Tixier-Durivault, 1955, now referred to Complexum, differ from C. pusillum sp. n. in lacking clubs in the surface of the polyparium. Moreover, both these species have many lobes and coloured polyps. Other encrusting Alcyonium species reported from the west Atlantic now referred to Complexum are A. patulum Tixier-Durivault, 1955 and A. strictum Tixier-Durivault, 1955, from Mauritania. A. patulum resembles A. strictum but is red with yellow polyps and has no clubs. Alcyonium strictum resembles A. pusillum sp. n. more than any other species, it also has clubs with a spiny head, but it has many small oval sclerites in the base, and is purple with yellow polyps.

A phylogenetic analysis has been performed, based on part of the mitochondrial COI gene and of the adjacent intergenic region (igr) which have been amplified according to McFadden et al. (2011). Additional Octocoral COI-igr sequences were retrieved from GenBank following a Blast search with the C. pusillum sequence as a query. As a comparison, Alcyonium spp. sequences were specifically retrieved from GenBank and included in the analysis. Alcyonium monodi sequences (referred to Complexum monodi) were kindly provided by Catherine McFadden and included in the dataset. They correspond to colonies sampled in 2012 in Senegal (10 km South of Dakar) at 15 m depth by Peter Wirtz. A phylogenetic reconstruction based on maximum likelihood (ML) has been performed with RaxML 8.1 (Stamatakis 2014) with a General Time Reversible + Gamma model and a rapid bootstrap analysis (1000 re-samplings). The nuclear gene coding for 28S ribosomal RNA has also been sequenced for C. pusillum following McFadden and Ofwegen (2013a) and the obtained sequence has been compared to other octocoral sequences thanks to a Blast search and a similar ML analysis. The COI-IGR sequence of C. pusillum has been deposited in GenBank (KP006396).

Tixier-Durivault (1955) described three other Alcyonium species from West Africa, A. altum, A. leave and A. violaceum. Verseveldt and Bayer (1988) referred A. altum and A. violaceum to the genus Nidaliopsis Kükenthal, 1906. It is unknown to us why Verseveldt and Bayer did not mention A. leave, we consider it to also belong to Nidaliopsis. This leaves Alcyonium senegalense Verseveldt & Ofwegen, 1992 from Senegal as the only Alcyonium species in West Africa. However, it has capstans and ovals in the coenenchyme (Verseveldt and Ofwegen 1992: figs 20–21), and therefore also does not match the current diagnosis of Alcyonium. We prefer to wait till molecular data of this species are available before describing another new genus to accommodate it. Superficially Complexum is similar to Alcyonium, it has the same type of colony shapes, and the sclerite arrangement in the polyps is also similar. Alcyonium differs in having clearly different sclerite types in surface layer and interior, radiates and club-like forms in the surface layer, and long spindles in the interior.

The phylogenetic analysis based on mitochondrial COI-IGR clearly confirmed the separation of Complexum from Alcyonium species (family Alcyoniidae) (Fig. 6). Indeed these species appear close to Eunicella (family Gorgoniidae; 99% bootstrap support) but the precise relationships between Complexum and Eunicella remain to be investigated with more species and the use of additional markers to get more robust information inside that group. A Blast analysis of the 28S sequence of C. pusillum in GenBank confirmed the closer relationship of this group with Eunicella compared to Alcyonium (data not shown). The current family-level taxonomy of octocorals needs extensive revision (McFadden et al. 2010) and therefore we keep the new genus in the family Alcyoniidae.

Figure 6. 

Unrooted ML phylogenetic tree based on mitochondrial COI-IGR with RaxML. Blue dots refer to groups on the right with more than 90% bootstrap support (over 1 000 bootstraps). Blue branches indicate Alcyonium species and red branches indicate Complexum species. *Alcyonium valdiviae has been transferred to the genus Parasphaerasclera by McFadden and Ofwegen (2013b). **Because of pre-occupation Alcyonium roseum has been renamed A. varum by McFadden and Ofwegen (2013b).