Borojevia tubulata sp.nov.

Figures 8a–c, 9a–f, 10a–h, 11a–f

Material examined. Holotype, RMNH Por. 10158, Maldives, Faafu Atoll, Wallstreet, 3.119°N 72.979556°E, depth 12 m, scuba, coll. N.J. de Voogd, field nr. MAD10 /MAS119, 20 February 2015.

Paratype, RMNH Por. 10113, Maldives, Faafu Atoll, Free Climbing, 3.066583°N 72.923028°E, depth 15 m, scuba, coll. N.J. de Voogd, field nr. MAD06 /MAS068, 18 February 2015.

Additional material, ZMA Por. 12435, Seychelles, Amirantes, Desroches Atoll, SW rim, outer reef slope, 5.7167°S 53.6167°E, depth 5–30 m, scuba, coll. M.J. de Kluijver, NIOP-E stat.nr. 774/04, 30 December 1992 .

Description. Irregularly arranged, groups of tubular individuals, the walls of which consist of tightly anastomosed tubuli. In situ white (Figs 8a, 10a) or cream or yellow-white (Fig. 8b) colored, on deck (Figs 8c, 10b) and preserved they are cream to pale beige (Figs 10c, 11a). Height of tubes up to about 3 cm, width 0.5 cm at the aperture and 1 cm at the base, but individual tubes may be smaller and thinner. Tube apertures are about as wide as the (pseudo-)atrium or slightly narrower. In preservation the tubes collapse to form easily damaged soft and flabby fragments (Figs 8c, 10b–c, 11a), which have their (pseudo-)atrium flattened to a narrow slit (Fig. 11b). The tube walls are ‘punctate’ (Figs 9a, 10d) caused by regularly distributed pseudopores (cf. Dendy’s (1913) description of Ascaltis gardineri, as Leucosolenia), which give access to the densely crowded tubuli.

Aquiferous system. Asconoid.

Skeleton. The pseudopores of the outer tube walls are strengthened by larger triactines, arranged in a distinct cortical pattern (Figs 9a, 10d), the skeleton of the walls of the tubuli are built by one-two layers of smaller triactines and tetractines, with the apical actines of the tetractines protruding into the lumina of the tubuli. The atrial sides of the tube walls (Fig. 9b) have an irregular skeleton and do not exhibit a special arrangement of larger triactines as observed on the outer wall, which indicates that the atrium is in fact a pseudoatrium, as is characteristic of the genus Ascaltis according to the Systema Porifera (Borojević et al. 2002a). See below for a further discussion.

Spicules. (Figs 9d–f, 10e–h, 11c–f) Triactines in two size classes, tetractines with spined apical actines. The latter invariably have the spines in three distinct rows the position of which match the basal triadiate system. We did not observe the presence of tripods, but we presume the larger category of triactines is homologous. Because of slight discrepancies between the specimens, we provide separate images of the samples and measurements of the three samples:

Holotype RMNH Por. 10158: Larger triactines (Fig. 9c) of the outer tube wall, equiradiate or very rarely sagittal, robust, with conical actines, which may slightly differ in length in the same spicule, actines 92– 133 –189 x 11 – 13.4 –16 µm.

Smaller triactines (Figs 9d,d 1), equiradiate (Figs 8d), or rarely sagittal (Fig. 9d 1) with undulate paired actines, measuring 54– 63 –111 x 5 – 6.3 –7.5 µm.

Tetractines (Figs 9e–f), with basal triradiate system usually equiradiate, but very occasionally there may be sagittal spicules, 51– 68 – 96 x 6 – 6.6 –9 µm, apical actines (Fig. 9f) 24– 37 – 48 x 3.5– 4.4 –5 µm.

Paratype RMNH Por. 10113: Larger triactines (Fig. 10e) with conical actines, 129– 151 –174 x 14 – 15.2 –18 µm; one single sagittal triactine was observed in the slides.

Smaller triactines (Figs 10f), with conical actines, 59– 72 – 93 x 6 – 7.2 –10 µm.

Tetractines (Figs 10g), with conical actines, those of the basal radiate system 48– 67 – 78 x 4.5– 6.6 –8 µm, apical actines with spines (Fig. 10h), 35– 43 – 58 x 4 – 5.2 –7 µm.

ZMA Por. 12435 (Figs 11c–f): Larger triactines (Fig. 11c), 102– 117.6 –134 x 11 – 13.4 –17 µm

Smaller triactines (Fig. 11d), 51– 75.1 – 94 x 6 – 8.4 –10 µm

Tetractines (Figs 11e), 61– 76.3 – 88 x 7 – 8.6 –10 µm, apical actines with spines (Figs 10f), 40– 53.4 – 74 x 6 – 6.6 –7µm.

Distribution and ecology. Maldives, Seychelles, under overhangs on coral reefs, depth 5– 30 m.

Etymology. Tubulatus (L.) means tube-shaped, referring to the habitus of the species.

Remarks. The three samples show some discrepancies, which we consider to be infraspecific variability. The two Maldives samples have similar habitus, but show some color differentiation between distinctly white and more creamy or pale yellow. The holotype has a larger proportion (though still a clear minority) of sagittal tri- and tetractines than the paratype. The Seychelles specimen was noted to be yellow in color, and it lacks sagittal spicules entirely. The Seychelles specimen has smaller triactines than the Maldives and the distinction between larger and smaller spicules is less obvious. For this reason, we limited the type material to that of the Maldives only. However, we believe that the three samples are conspecific.

The species is assigned to the genus Borojevia despite the elaborate shape and presence of a pseudoatrium, and despite the absence of clear tripods among the triactines. In fact, we initially assigned the specimens to the genus Ascaltis, based on the habitus and skeletal structure and because a Western Indian Ocean species, Ascaltis gardineri (Dendy, 1913) appeared to be close in skeletal characters. The major difference between that species and our specimens is the lobate, non-tubular habitus in the former.

Molecular sequence data (partial 28S, cf. Fig. 2B) were obtained for the holotype and the Seychelles material of Borojevia tubulata sp.nov. Our phylogenetic analysis of Calcinean sequences, obtained from our own Western Indian Ocean samples, Voigt et al. ’s (2017) samples (we were allowed to include a recent sequence of Borojevia sp. from the Maldives kindly sent to us by Oliver Voigt, which is likely a member of the present new species), and from GenBank, clearly showed that the present species is closely related to Borojevia aspina (Klautau et al., 1994), Borojevia aff. aspina sensu Voigt et al. 2017, B. cerebrum (Haeckel, 1872), B. brasiliensis (Solé-Cava et al. 1991) and a new Borojevia species decribed below, and was distant from Ascaltis reticulum (Schmidt, 1862), the only Ascaltis for which molecular data are available. As Klautau et al. (2016) pointed out, we cannot be certain about the molecular affiliation of the genus Ascaltis yet, because there are no sequences available for the type species Ascaltis lamarcki (Haeckel, 1872) . We have chosen to follow the molecular signal and keep our species in the genus Borojevia, against the weight of morphological evidence for a possible membership of Ascaltis .

The new species differs from the above-described Borojevia voigti sp.nov. in the presence of a pseudoatrium and the more prominent and consistent spination of the apical actines of the tetractines.

A further similar species appears to be Leucascus simplex Dendy, 1892, from South Australia, reported also from Providence Island in the Seychelles by Dendy (1913). The type of that species has spined apical actines (cf. redescription of L. simplex in Cavalcanti et al. 2013), but it has not been verified for the Providence specimen. This differs from the present species also in habitus (massive, with only a slit-like atrial cavity). Below we describe a new species of Leucascus from Eastern South Africa. Apart from a more elaborate shape, it is distinct by having a surface palisade of short diactines. The present species is not a likely Leucascu s, as in our phylogeny of Fig. 2, Leucascus flavus Cavalcanti et al,. 2009 is not at all closely related to Borojevia species.