Geodia phlegraei, , Koltun, 1964

Cárdenas, Paco, Rapp, Hans Tore, Klitgaard, Anne Birgitte, Best, Megan, Thollesson, Mikael & Tendal, Ole Secher, 2013, Taxonomy, biogeography and DNA barcodes of Geodia species (Porifera, Demospongiae, Tetractinellida) in the Atlantic boreo-arctic region, Zoological Journal of the Linnean Society 169 (2), pp. 251-311 : 284-293

GEODIA PHLEGRAEI ( SOLLAS, 1880b)

GEODIINAE P PHLEGRAEI (PhyloCode SPECIES NAME)

( FIGS 18–20 , 23 , TABLE 5)

Isops phlegraei, Sollas, 1880b: p. 397 ; 1888: p. 267; von Lendenfeld, 1903: p. 95; Hentschel, 1929: p. 919; Burton, 1930: p. 498; Vosmaer, 1933: p. 141; Alander, 1942: p. 73; Warén & Klitgaard, 1991: p. 52; Klitgaard, 1995: p. 2; Reitner & Hoffmann, 2003: table 1; Klitgaard & Tendal, 2004: p. 57; Cárdenas et al., 2010: p. 91.

Synops pyriformis, Vosmaer, 1882: p. 20 ; 1885: p. 10 (synonymy by Burton, 1930).

Isops sphaeroides, Vosmaer, 1882: p. 13 (synonymy by Burton, 1930).

Isops pallida, Vosmaer, 1882: p. 16 (in part); Rezvoj, 1928: table 1 (synonymy by Burton, 1930).

Isops pyriformis, von Lendenfeld, 1903: p. 95 .

Geodia phlegraei, Koltun, 1964 (in part): p. 148; Koltun, 1966 (in part): p. 55; Cárdenas et al., 2011 (in part): table S1; Murillo et al., 2012 (in part): p. 842; Schöttner et al., 2013: p. 2.

Not:

Isops pyriformis, Hentschel, 1929: p. 865 (= G. parva ).

Geodia cf. phlegraei, Wagoner et al., 1989: p. 2344 (= G. parva ).

Isops phlegraei pyriformis, Klitgaard & Tendal, 2004: p. 57 (= G. parva ).

Isops phlegraei, van Soest et al., 2007: p. 129 (= G. cf. nodastrella ).

Geodia phlegraei, Koltun, 1966 (in part): p. 55; Cárdenas et al., 2011 (in part): table S1; Murillo et al., 2012 (in part): p. 842 (= G. parva ).

Misidentifications:

Pachymatisma johnstonia, Hansen, 1885: p. 17 .

Cydonium mülleri, Breitfuss, 1912: p. 62 (according to Koltun, 1966).

Geodia cydonium, Gorbunov, 1946: p. 37 (according to Koltun, 1966).

Type locality and type material examined

Isops phlegraei, Korsfjord near Bergen , Norway. 60°10′N, 05°10′E, 330 m. Collected by Rev. A. M. Norman in 1878, BNHM 1910.1 .1.840.

Synops pyriformis , near Hammerfest , Norway, 71°12′5″N, 20°30′5″E, 247 m, Willem Barents Exp. 1878–79, RMNH Por. 660 (wet specimen); MNHN- DN23, Norman Collection, spicule preparation and section; BMNH 10.1.1.1147–1148, Norman Collection, spicule preparation and section (not seen) .

Isops sphaeroides , near Hammerfest, Norway, 71°12′5″N, 20°30′5″E, 247 m, Willem Barents Exp. 1878–79, MNHN-DN24 , syntype 1, Norman Collection, spicule preparation and section ; RMNH, Vosmaer slide collection, box number 37, syntype 2, two spicule preparations and one stained thin section ; BMNH 10.1.1.1158–1159, Norman Collection, two slides (not seen).

Isops pallida , near Hammerfest, Norway, 71°12′5″N, 20°30′5″E, 247 m, Willem Barents Exp. 1878–79, RMNH Por. 652, wet specimen (seen only from pictures); RMNH, Vosmaer slide collection, box number 37, two spicule preparations with number 62.

External morphology and cortex: More or less spherical, the largest specimens somewhat flattened on the top, cup-shaped, generally up to 20 cm in diameter, and 15 cm high ( Fig. 18A, B, E ); our largest specimen measures 43 cm in maximum dimension. Juveniles are spherical ( Fig. 18C ). Specimens growing on vertical cliffs (i.e. fjords) can be flattened or more irregular ( Fig. 18D ). Lower part sometimes formed as a base, with root-like outgrowths ( Fig. 19B ). The colour alive usually varies from whitish grey to light brown, to slightly rose. One shallow fjord specimen (PC111, shallower than 175 m) was faintly purple in ethanol, like shallow G. barretti and Pachymatisma normani Sollas, 1888 (a NEA boreal Geodiidae ) ( Cárdenas & Rapp, 2013). Other specimens fixed in ethanol seemed to be reddish, although this staining comes from the epibiont yellow sponge Hexadella dedritifera Topsent, 1913 ( Fig. 18A, F ) which turns dark red during the fixation. Many specimens are very hispid on the sides, but never on the top surface ( Fig. 18A–C ). The fur can be up to 10 mm long, but is not a regularly occurring feature in dredged specimens as it may be easily lost during the collection. The cortex ( Fig. 18I ) is thin to fairly thick (0.7–2 mm thick); it is usually difficult to cut and tends to break in pieces. Many epibionts (especially sponges) are present in this hispid part. Uniporal oscular openings are up to 1 mm in diameter, and are found mainly on the upper surface ( Fig. 18A– F ). Oscular openings may be at the tip of thick conical elevations, more or less pointy, which can be lighter-coloured ( Fig. 18G ) (but not always) and which gave the name to this species: Campi phlegraei is a local name of the Naples volcanic area ( Sollas, 1880b). Uniporal pores (c. 300–400 Mm in diameter) ( Fig. 18H–-J ) are scattered on the sides and partly on the underside of the body. Pores are usually not elevated but can also be surrounded by a white margin.

Description of type material: The holotype of G. phlegraei is a small subspherical specimen (diameter: 2.5 cm) cut up into four parts ( Fig. 19A ); it has conical-shaped oscules. There is also one BMNH slide

Means are in bold; other values are ranges; N = 30 unless stated otherwise in parentheses, or unless measurements come from other studies. A dash indicates that this measurement is not given in the literature. n.f., not found; n.o., not observed in the specimen in our possession. N, Norway; S, Sweden.

of the type but it is damaged and the embedding medium has blackened. Figure 20 shows SEM pictures of the spicules from the holotype. The type of S. pyriformis is a medium-sized specimen (length: 10 cm, width: 8 cm) cut into five pieces. The main piece ( Fig. 19B ) is the elongated cup-shaped half represented by Vosmaer (1882: plate IV). We have only seen pictures of the wet specimen of the lectotype (specimen a) of I. pallida ( Fig. 19C, D ): the pearshaped specimen is cut into two fragments (one-half and one-third of a single specimen). The half fragment is about 6 ¥ 4 cm with a thick cortex (1–2 mm thick), uniporal oscules, and uniporal pores. There are also two spicule preparations of the lectotype (slide 62: a) of I. pallida ( Fig. 4A ). One slide is broken so the label is gone but the spicules are identical to those of slide 62.

Spicules ( Fig. 20 , Table 5): Megascleres: (a) oxeas, length: 1173–7600 Mm; width: 10–100 Mm. (b) Orthotriaenes, rare dichotriaenes, rhabdome length: 586– 6655 Mm; width: 12–150 Mm; orthotriaene clad length: 80–1125 Mm; protoclad length: 220–250 Mm; deuteroclad length: 100–250 Mm. (c) Anatriaenes, rare (some clads were dichotomized in the type of I. sphaeroides ) rhabdome length: up to 11 mm; width: 8–25 Mm; clad length: 30–130 Mm (minimum according to Koltun, 1966). (d) Protriaenes, very rare, rhabdome length: 11 600–12 750 Mm; width: 25–30 Mm; clad length: 190–200 Mm. Microscleres: (e) sterrasters, subspherical (NEA specimens) or spherical (in some NWA specimens), length: 82–144 Mm; width: 70–124 Mm; thickness: 65–80 Mm; hilum: 12–15 Mm. Rosettes are made of 4–12 smooth rays; rosette diameter: 6–8 Mm. (f) Spherasters with spiny conical actines (more rarely with blunt ends), 8–26 Mm in diameter. (g) Oxyasters, smooth (rough actines were rarely observed in very large oxyasters), 10–70 Mm in diameter (maximum is from measurements of the type by Sollas, 1888).

DNA barcodes: GenBank accession nos. EU442196 , HM 592701 (Folmer COI). We have sequenced COI from specimens from Spitsbergen (1), western and northern Norway (10), Mingulay Reef (1), and Orphan Knoll (1): the Folmer COI is identical in all these specimens. No. KC481222 (18S), obtained from ZMBN 77929 (Korsfjord, Norway). 18S of ZMBN 89719 (Spitsbergen) was also sequenced: no variation was observed .

Distribution ( Fig. 23 ): Geodia phlegraei has a boreal distribution and seems to avoid arctic waters; it can be found from 40 m (Trondheimsfjord) to 3000 m (Orphan Knoll). It is commonly found at depths from 100–300 m (Norwegian continental shelf) to 725 m (Faeroes), at temperatures of 0.3 °C (BIOICE, st. 2926) to 7.9 °C (BIOFAR, st. 297). It has also been recorded by divers in Norwegian fjords at shallower depths: for example, it has been photographed by A. Salesjö in the Trondheimsfjord at only 40–50 m depth (http:// www.anderssalesjo.com/?id=3306&lang=42, accessed 7 May 2013). We also identified it in material collected in Mingulay Reef (western Scotland) at 128–139 m depth. In Mingulay reef (R. van Soest, pers. comm.) and Norwegian fjords at shallow depths, temperatures can reach 10.5 °C. The specimen identified as G. phlegraei from Rockall Bank ( van Soest et al., 2007) was actually a Geodia cf. nodastrella . Interestingly, all but one of the NWA specimens from the Flemish Cap that we examined (a total of eight specimens) collected during NEREIDA 2009–10 and originally identified as G. phlegraei were in fact G. parva . The only specimen from the Flemish Cap that we identified as G. phlegraei (DR24-69d = UPSZMC 78280) has an external morphology similar to G. phlegraei and large elongated sterrasters (c. 132–136 Mm in diameter). Unfortunately, we could not get a COI sequence from it to confirm this identification. However, we did get a COI sequence for R1340-04 (= UPSZMC 78308) confirming it was G. phlegraei : it has large spherical sterrasters (88–107 Mm in length) and it was collected at 3000 m depth at Orphan Knoll, where the temperature was 2.4 °C ( Fig. 18F shows R1341-18 collected at 2347 m in the same area). The most western specimen of G. phlegraei found is from western Greenland (‘Shinkai Maru’, st. 32, 64°13.5′N, 54°42.1′W. 970 m) ( Fig. 18G, H ), but this material was not suited for molecular studies.

Biology: When he described this species, Sollas (1880) immediately noticed that it was ‘covered by various foreign bodies’. Indeed, G. phlegraei is the boreoarctic Geodia which is most often found covered with epifauna, especially sponges and including individuals of its own species. Here are a few sponges found on G. phlegraei : Craniella sp. , Cyamon spinispinosum ( Topsent, 1904) , Hexadella dedritifera (mis-identified as Aplysilla sulphurea in Klitgaard (1995)), Leucandra spp. , Lissodendoryx (L.) fragilis Fristedt, 1885 , Polymastia grimaldii ( Topsent, 1913) , Stelletta normani , Ute gladiata Borojevic, 1966 , etc. G. phlegraei can settle on other sponges as well such as other large Astrophorina [ S. normani , Stryphnus fortis ( Vosmaer, 1885) ]. The predator chiton H. nagelfar and the parasitic foraminiferan H. sarcophaga have been found living on G. phlegraei ( Warén & Klitgaard, 1991; Cedhagen, 1994; Todt et al., 2009). More associated fauna has been investigated by Klitgaard (1995). The chemistry (elemental analysis, amino acids, sterols, and quaternary ammonium compounds) has been investigated by Hougaard et al. (1991a, b).

The associated microsymbionts of G. phlegraei collected in the Sula Ridge reef ( Norway) have been studied ( Graeber et al., 2004; Dieckmann et al., 2005) and led to the isolation and description of a new gammaproteobacterium ( Oceanospirillales group): Spongiispira norvegica ( Kaesler et al., 2008) .

We observed on the type section of I. sphaeroides made by Vosmaer many subglobular oocytes without pseudopodes. This specimen has been collected in the Barents Sea on 2 July 1879.

In the NEA, G. phlegraei can easily be confused with another Geodiidae , Pachymatisma normani , which also has raised white-rimmed uniporal oscules, but P. normani has microrhabds in the cortex, instead of spheroxyasters. Geodia phlegraei can also be easily confused with its sister species G. parva , and in that case only spicule and genetic characters can differentiate them (cf. below).

Distinctive characters: External morphology: round to sometimes cup-shaped, with only uniporal openings (i.e. no sieve). The numerous small oscules on the top of specimen, each with a whitish rim making it look like a little wart. Often overgrown with other sponges, hydrozoa, bryozoa, etc. Spicules: large spherasters, smooth oxyasters along with fairly large sterrasters (70–144 Mm).

Remarks: Choanosomal oxyasters can sometimes be separated into two size categories (10–25 and 45–60 Mm) but in some specimens we have more of a continuum of sizes so we decided to treat oxyasters as one category ( Table 5). Koltun (1966) notes the presence of rare small slightly curved oxeas (230–420 ¥ 8–10 Mm); we never observed those and wonder whether these could have been contamination. Dichotriaenes have been reported by Koltun (1966) and Vosmaer (1882), but it should be stressed here that they are fairly rare as we only found a few (notably in the holotype). Anatriaenes are rare and we report for the first time the presence of protriaenes in this species (found in the type of I. sphaeroides ), but they seem to be very rare.

Geodia phlegraei is the type species of Isops Sollas, 1880b , a genus synonymized with Geodia based on molecular results ( Cárdenas et al., 2010). Synops pyriformis is the type species of Synops , a genus later synonymized with Isops ( Sollas, 1889) . Vosmaer’s (1882) species ( Isops sphaeroides , Isops pallida , Synops pyriformis ) were all described from the same station near Hammerfest (71°12′5″N, 20°30′5″E) and collected on the same day (2 July 1879). Burton (1930: 498) writes: ‘Examinations of the preparations in the Norman Collection, made from the type-specimens of these species, show conclusively that Geodia parva Hansen , Isops pallida Vosmaer , I. sphaeroides Vosmaer , and Synops pyriformis Vosmaer are synonymous with Isops phlegraei Sollas. ’ We stated earlier that Burton did not look at the type of I. pallida , and that a lectotype ( Fig. 19C, D ) was designated to keep the synonymy with G. phlegraei . On the other hand, Burton did look at the BMNH type slides of I. sphaeroides and S. pyriformis .

Isops sphaeroides was described from two globular syntypes (both presumably lost, R. van Soest & N. de Voogd, pers. comm.) : one larger specimen was hispid, whereas the smaller was not. All type slides we examined were identified as G. phlegraei ( Table 5). However, we found differences between the MNHN type slides ( Norman Collection ) and those from the Vosmaer slide collection, namely differences in cortex thickness and sterraster sizes ( Table 5), so these may represent the two syntypes. The Vosmaer slides include a section from half of a smooth specimen with an estimated total diameter of 2–2.5 cm, which must be the smaller syntype . To conclude, we presume that the Vosmaer slides correspond to the smaller syntype so the MNHN slides correspond to the larger syntype .

We also compared type specimen and type slides of S. pyriformis with our material. The cortex thickness of S. pyriformis (1.1–1.3 mm) agrees well with G. phlegraei ( Table 5). Spicule measurements also match those of G. phlegraei ( Table 5). Finally, observation of the external morphology ( Fig. 19B ) and new spicule preparations from the wet type of S. pyriformis further confirmed this. All in all, we therefore follow Burton (1930) and confirm the synonymy for S. pyriformis , I. pallida , and I. sphaeroides .

Koltun (1966) had already underlined the morphological variability of G. phlegraei . He notably states that in deeper and colder waters the morphology of G. phlegraei is somewhat different. Klitgaard & Tendal (2004) also noticed this arctic water morphotype and considered it as a subspecies of G. phlegraei : G. phlegraei pyriformis . After having examined many specimens from the whole boreo-arctic area, we confirm the existence of different morphotypes, and, after incorporating additional morphological data as well as molecular data, it was decided that the arctic morphotype represented a valid species which had in fact been previously described under the name Geodia parva Hansen, 1885 , before being synonymized with G. phlegraei ( Burton, 1930) . Below, we resurrect and redescribe G. parva .

Figures

Figure 4. A, type slides of Isops pallida Vosmaer, 1882 for specimen numbers 62 (= a, here designated as lectotype) identified as G. phlegraei, and 64 (= b) identified as G. atlantica. B, orthotriaenes and abundant anatriaenes from one of the slides 64 (= b). C and D, thick sections of the holotype of Sidonops atlantica Stephens, 1915, specimen S.R.151-27/ 364–1914. E and F, thick sections of ZMBN 77927, Korsfjord, 200–400 m depth. Scale bars: B, 400 Mm; C, E, F: 500 Mm; D, 300 Mm.

Figure 18. External morphology and thick sections of Geodia phlegraei Sollas, 1880b. A, specimen PC124 just after dredging from the Korsfjord, Norway, 200–400 m depth; specimen is 20 cm in diameter. B, specimen ZMBN 85211 just after collecting from Traenadjupet, northern Norway, 290 m depth: specimen is 20 cm in diameter. C, specimen PC68 just after dredging from the Korsfjord, Norway, 200–400 m depth; specimen is 4 cm in diameter. D, specimen PC191 just after dredging from the Korsfjord, Norway, 200–400 m depth; specimen is 20 cm in length. E, specimen from Ingolf Exp., st. 90, Denmark Strait, 1039 m depth (fixed in ethanol); specimen is 15 cm high and 8 cm in diameter. F, specimen R1341-18 in situ, collected at 2347 m depth, Orphan Knoll (50.05894, 45.60638); specimen is 20 cm in length and covered with the yellow encrusting sponge Hexadella dedritifera (Canadian DFO/ROPOS 2010). G, uniporal oscules (specimen from Shinkai Maru Exp., st. 32). H, uniporal pores (specimen from Shinkai Maru Exp., st. 32). I, thick section from ZMBN 77929 (Korsfjord, 200–400 m). J, uniporal pore with sphincter. Scale bars: G: 5 mm; H: 1 cm; I: 1 mm; J: 200 Mm.

Figure 19. A, holotype of Geodia phlegraei SOLLAS, 1880b, BMNH 1910.1.1.840. B, holotype of Synops pyriformis Vosmaer, 1882, RMNH Por. 660. C, lectotype of Isops pallida Vosmaer, 1882, RMNH Por. 652, external appearance (picture: R. van Soest). D, lectotype of I. pallida, RMNH Por. 652, internal appearance (picture: R. van Soest).

Figure 20. Spicules of the holotype (BMNH 1910.1.1.840) of Geodia phlegraei Sollas, 1880b. A, oxyasters. B, spherasters. A and B have the same scale. C, orthotriaene. D, sterrasters. E, close-up on the hilum of a sterraster. Note the smooth rosettes. Scale bar: 3 Mm.

Figure 23. Distribution of Geodia phlegraei Sollas, 1880b (dark dots) and Geodia parva Hansen, 1885 (light dots). White dots represent specimens of G. phlegraei or G. parva but that we could not assign to the proper species for various reasons (map made with GeoMapApp, http://www.geomapapp.org). T, type locality.