Paraleucilla dalmatica Klautau et al. 2017 P. incomposita Cavalcanti et al. 2014 Paraleucilla P. magna Klautau et al. , 2004 P. erpenbecki Paraleucilla Calcareous sponges of the Western Indian Ocean and Red Sea Van, Rob W. M. De, Nicole J. Zootaxa 2018 2018-06-01 4426 1 1 160 4CYRP Van & De, 2018 Van & De 2018 [151,438,1448,1474] Calcarea Amphoriscidae Paraleucilla GBIF Animalia Leucosolenida 133 134 Porifera species erpenbecki sp. nov.   Material examined. Holotype, ZMAPor. 22409a, MozambiqueChannel, between Mozambiqueand Madagascar, E of Juan de Nova Island, 17.2817°S 43.1567°E, depth 60 m, coll. RV ‘Pelagia’ Around Africa II expedition, leg 6, field nr. 20- ASC 10, 1 April2001  Paratypes, ZMAPor. 22409c, five individuals from the same locality as the holotype.   Description.The sample consists of six sycon-like tubular individuals ( Figs 83a,a 1), one of which was chosen as the holotype( Fig. 83a). Shape oval, narrowing towards the osculum. Height of largest tube (the holotype) is 3 cm, diameter 1.5 cm. Fringe prominent but short, 2–3 mm, slightly flaring. Color in alcohol light beige. Surface slightly hispid, rough-looking. Consistency soft.  Aquiferous system. Leuconoid.  Skeleton.( Figs 83b–d) The wall has a thickness of about 2 mmwith protruding trichoxeas and very few diactines causing the hispid surface. In the SEM cross section ( Fig. 83b), from the periphery towards the atrium, there is a cortical skeleton ( Fig. 83c) of rare triactines carried by the basal actines of a single layer of cortical tetractines. Occasionally, there are scattered diactines protruding from the skeleton to the outside. Next, the choanosomal skeleton is inarticulate ( Fig. 83b) formed by the apical actines of the cortical tetractines and unpaired actines of giant triactines lying in the mid-region of the wall. Below these there is a confused mass of smaller tetractines and triactines, and finally the atrial skeleton ( Fig. 83d) is formed by tetractines of which the apical actines protrude far into the atrial lumen, and by smaller triactines. The fringe is formed by thin giant diactines grading into thick trichoxeas, and at the base it is supported by triactines and tetractines (not shown).  Spicules.( Figs 84a–i) Giant diactines, trichoxeas, small diactines, large tetractines, large triactines, small tetractines and small triactines. Giant diactines (Fg. 84a), fusiform, blunt endings, 388– 1066– 1973 x 19– 31.8–42 µm. Trichoxeas and thin diactines ( Fig. 84b), sharp endings but subapically often slightly distended, almost invariably broken, fragements measuring 480– 962– 1740 x 3– 4.8–7 µm. Small diactines ( Fig. 84c), not common, only a few could be measured, 147–210 x 8–9µm. Cortical triactines and small triactines of the subatrial region ( Figs 84d), these were indistinguishable and not very common, either regular or slightly irregular or sagittal, actines 138– 196–266 x 8– 10.9–16 µm. Cortical large tetractines ( Fig. 84e), not very common, with basal actines often curved and blunt ending, apical actines straight and pointed, thinner than the other actines; unpaired actines 132– 324–468 x 13– 30.5–37 µm, paired actines 216– 304–429 x 15– 21.7–31 µm, apical actines 211– 321–786 x 18– 17.1–23 µm. Giant triactines ( Figs 84f), sagittal with straight unpaired actines and curved paired actines; unpaired actines 151– 320–603 x 10– 20.4–33 µm, paired actines 231– 366–598 x 12– 22.2–36 µm. Giant tetractines of the subatrial region ( Fig. 84g), with straight actines, apical actines short and conical; unpaired actines 231– 394–696 x 13– 18.8–38 µm, paired actines 228– 313–391 x 12– 16.7–26 µm, apical actines 45– 62– 84 x 7– 8.3–10 µm. Small tetractines of the subatrial region ( Fig. 84h), similar in shape to giant tetractines, but smaller, unpaired actines 48– 149–268 x 7– 10.4–13 µm, paired actines 63– 151–249 x 6– 9.0–11 µm, apical actines 31– 47– 61 x 5– 7.2–10 µm. Atrial tetractines ( Fig. 84i), with long straight apical actines and unpaired actines almost similar in length, with curved paired actrines; unpaired actines 152– 232–418 x 9– 11.3–13 µm, paired actines 174– 285–461 x 8– 9.1–12 µm, apical actines 66– 167–234 x 8– 8.9–10 µm.   Distribution and ecology. MozambiqueChannel, at 60 mdepth.   Etymology.Named after Dr. Dirk Erpenbeck, München, Germany, in recognition of his great efforts to integrate molecules and morphology in the classification of the Porifera.   Remarks.The new species is assigned to  Paraleucillaon account of the skeletal zonation of an inarticulate subcortical skeleton formed by the apical actines of the cortical tetractines and the unpaired actines of giant triactines in the mid region of the choanosome, followed by a confused choanosomal and subatrial skeleton. A regional species is  Paraleucilla proteus(Dendy, 1913)(originally as  Leucilla). It has the same shape (although much smaller: only 7 mmhigh) as our new species. Differences are that almost all spicules, excepting the subcortical tetractines, are considerably smaller in size, notably the giant triactines do not seem to be represented, the apical actines of the atrial tetractines are much shorter, and there are no giant diactines (though smaller diactines are present). There are also no cortical small triactines. Dendy suggested that his small specimens were juveniles of the Australian species  Leucilla australiensis( Carter, 1886), but eventually Borojević et al.(2000) assigned Dendy’s species to  Paraleucilla. A further geographically close species is  Paraleucilla cucumis( Haeckel, 1872)(p. 205, as  Leucandra) from Sri Lankaand South Australia, differing a.o. in the presence of subcortical andmid-region tetractines (the latter not present in our new species), and the absence of mid-region and atrial triactines, with as a consequence the absence of apical actines protruding into the atrial lumen. There is a small nomenclatorial problem, because Hackel (1872)divided his  Leucandra cucumisinto two (?) varieties (‘spezifische Varietäten’),  L.c.var. bassensisand  L.c.var. palcensis. The difference was the virtual absence ( bassensis) and presence ( palcensis) of giant diactines. The difference was apparently considered trivial because neither Dendy (1892)and Dendy & Row (1913), nor Cavalcanti et al.(2014)make mention of these varieties. We formally need to indicate which one of the varieties is the nominotypical variety (ICZN art. 47). In view of the redescription by Dendy (1892)it makes sense to consider the var. bassensisas the typical variety, to be named  Leucandra cucumisvar. cucumis, with the var. palcensisas a junior synonym. The latter name would then be available if future research would result in distinction of an Indian Ocean species differing from the Bass Strait  Paraleucilla cucumis.   FIGURE 83.  Paraleucilla erpenbecki  sp.nov., a, habitus holotype ZMA Por. 22409a, from the Mozambique Channel, a1 habitus of paratypes ZMA POR. 22409c, from the same location (scale bar = 1 cm), b–d, SEM images of skeleton, b, cross section, c, overview from above of peripheral skeleton, d, overview from above of atrial skeleton. The new species is close to West Australian  Paraleucilla princeps( Row & Hôzawa, 1931)(as  Leucilla princeps, p. 799, pl. 21 fig. 17, text-fig. 16), sharing tubular shape and most of the spicule complement. Also the sizes of the spicules conform rather closely. A major difference is the absence of giant triactines in the choanosomal skeleton. These spicules are a dominant feature of the present species, forming the skeletal structure of the peripheral inarticulate skeleton together with the apical actines of the subcortical tetractines. Row & Hôzawa do mention the presence of similar shaped smaller triactines but these only occur in the oscular region. Also not present are the small diactines.    Paraleucilla dalmatica Klautau et al.2017from the Mediterranean is similar in shape and skeletal structure, but spicule sizes differ significantly. Also, Brazilian  P. incomposita Cavalcanti et al.2014is close in shape and structure but differs also in spicule sizes. Recently ( Chagas & Cavalcanti 2017), it was discovered that this species possesses choanosomal pentactines as a remarkable unique feature. We obtained a 28S partial gene sequence for the holotypeand compared it with a sequence downloaded from GenBank of Mediterranean  Paraleucillaspec. (supposedly the same as  P. magna Klautau et al., 2004), which grouped together in our Phylogeny of Fig. 3at a moderate bootstrap frequency (77%). From a separate investigation of a trimmed alignment of 295 sites obtained for the two sequences we found that  P. erpenbecki  sp.nov.differs in 8 sites from the Mediterranean  Paraleucillaspec.