Sabellastarte sanctijosephi (Gravier, 1906)

Knight-Jones, Phyllis & Mackie, Andrew S. Y., 2003, A revision of Sabellastarte (Polychaeta: Sabellidae), Journal of Natural History 37 (19), pp. 2269-2301 : 2292-2296

Sabellastarte sanctijosephi (Gravier, 1906)

(figure 9)

Eurato sanctijosephi Gravier, 1906 (plate 7, figures 281–283), 1908 (text), Djibouti (MNHNP A258, A312 and A469 original material).

Sabellastarte sanctijosephi: George and George, 1979 , figure 58.4; Vine, 1986: figures 82, 88. [Not Day, 1967, Gibbs, 1969, nor Bailey-Brock, 1976, 1987, all of which are S. spectabilis .]

Sabellastarte indica: Fauvel, 1955 , Abulat, Red Sea; 1958, Gulf of Aqabar; Schuhmacher and Hinterkircher, 1996: 1179 (top left figure).

Additional material. Suez, one specimen ( ZMH PE 1345 ); Gulf of Aqabar , Elat , one specimen from a Naval sublittoral ‘pillar’, 100 m N of Deckel Beach (collected for P.K.J. by Aharon Meroz); Israel Red Sea Expedition, Dahlak Archipelago ( ZMA V. Pol. 1271, 10057, 10076, 10100, 10101, 1271 as S. indica ) .

The material described here is one of Gravier’s (1906, 1908) largest and most complete specimens from Recife du Meteore, Djibouti (MNHN A258 vial ‘e’, two specimens), which had Gravier’s original ‘Golfe de Tadjourah’ label, and another, inserted by Fauvel in 1940, re-identifying it as Sabellastarte indica ; MNHN A258 also contains five other vials with Gravier’s original labels with two, three, one, one and one specimens (some also relabelled S. indica by Fauvel) . MNHN A312 and A469 also have original labels noting Golfe de Tadjourah or just Djibouti and just one vial in A312 has the additional information, Isle Musha. The longest specimen noted by Gravier (1908) was 85 mm long (without crown), 8 mm wide, crown about 33 mm long. Data in parentheses refer to the live specimen studied by P.K.J. in Elat .

Description. Body without crown 60 (82) mm long, 9 (8.4) mm wide with about 115 segments of which seven to eleven, usually eight, are thoracic (figure 9A); crown about 40 (38) mm long with a fairly short base involuted ventrally to form a semicircle each side with D-shaped flange on each dorsal margin (figure 9D) below the interadiolar web region, web fairly long equal to length of about three thoracic shields, radioles about 42 each side, without interdigitation, outer surface of rachis subquadrangular (figure 9E, F, G), tips beyond pinnules short and blunt (figure 9H); dorsal lips attenuated, about 15 mm long, grooved, with midrib support and webbed to adjacent radiole without pinnular support; thorax wider than long (measured laterally), last thoracic torus about one-third length of longest (second torus), first segment about 1.5 times length of each following thoracic segment (viewed laterally and discounting height of collar); dorsal-most collar margins starting at thorax/crown junction, poorly developed (not forming lappets), extending posteriorly towards first fascicle, forming U-shaped emarginations above collar pockets (figure 9D) exposing peristome, dorso-lateral margins extending sharply anteriorly to well above thorax/crown junction; ventral collar forming two subtriangular lappets which may project forward to overlap at midline (figure 9B); first ventral shield longer than others in thorax, embayed medially with rest of margin sloping posterolaterally (figure 9B); all thoracic tori abutting lateral margins of ventral shields; tube, chaetae, uncini and length of thoracic fascicles similar to those of S. spectabilis ; thoracic fascicle 13 with about 70 chaetal shafts (many broken distally), 13th abdominal fascicle with 31 chaetae (11 superior and 20 inferior); interramal eyes minute in abdomen (often camouflaged by background flecks), larger in thorax.

Gravier (1908) recorded bands of violet alternating with colourless zones around the crown. These bands have now faded to brown and light brown. The background body colour is also light brown, speckled with small medium or dark brown flecks, which tend to coalesce into darker brown areas between the uncinal ridges, over anterior ventral shields and along the crown base as axial stripes. The ventral shields and glandular areas between them and the parapodia are similar in colour, i.e. light to medium brown.

Variation. Some specimens that have been preserved in their tubes may appear to lack the shallow dorsal part of the collar flanking the midline groove, so that the dorso-lateral parts extending anteriorly appear to have dorsal margins separated by a wide gap. Inserting a probe can establish the presence of the shallow collar pockets flanking the dorsal midline groove.

Light pigment bands are present on crowns of specimens from the Sudanese Red Sea (Sanganeb lagoon, Vine, 1986: 89), but absent on crowns from Elat. The specimen collected at Elat (P.K.J.) had radioles with pale yellow pinnules throughout and a brown line along outer surface of each rachis. Of four specimens in the Elat aquarium, one crown was yellow, another white and two orange. About 10 other specimens, left undisturbed amongst a domed coral outcrop (10–12 m diameter) near the Egyptian border, were all orange. Vine (1986: 82, 88 bottom) shows two pale, unbanded specimens from the northern Egyptian Red Sea , with the brown stripe along each rachis as in the live specimen from Elat. The ventral shields were salmon-pink (now straw-coloured) and the rest of the body was greyish pink and flecked .

As Gravier (1908) noted, irregularities occur in the disposition of chaetae in both thorax and abdomen. These irregularities are due to imperfect regeneration after damage.

Habitat. Found in the shallow sublittoral, common amongst coral, but as the Elat specimen was on a fauna-encrusted iron pillar, the species will presumably thrive in a sheltered spot on any hard substratum. Gravier (1908) notes that it was not found beyond 20 m depth.

Remarks and distribution. Gravier placed his species in Eurato , a genus erected by Saint-Joseph (1898: 249) for species like Sabellastarte , but with radioles not interdigitating to form two concentric semicircles. None of his included species (footnotes, pp. 249, 250) belong to Sabellastarte . The first mentioned species is Sabella pyrrhogaster Grube (1878) , designated type species of Eurato by Bush (1905). Sabella pyrrhogaster is a species of Notaulax Tauber (1879) and Eurato is therefore a junior synonym of Notaulax (Perkins, 1984) .

It has been a common mistake to regard Sabellastarte sanctijosephi as a ‘young’ form of S. indica (now S. spectabilis ). Even the larger S. sanctijosephi have numerous radioles without interdigitation, in situ arranged into a smooth bowl-shape each half with a small ventral involution, as in George and George (1979: 58, figure 4), Vine (1986: 88) and Schumaker and Hinterkircher (1996: 116, top left as S. indica ). Those with interdigitating radioles like Sabellastarte spectabilis form mop-like crowns in situ, as shown by Gibbs (1969, figure 137 as S. sanctijosephi ), George and George (1979: 58, figure 6, as S. sanctijosephi ) and Fossa and Nilsen (1996: 130, 131 as S. indica , and 2000: 130, 131 as S. spectabilis ). Furthermore, Sabellastarte sanctijosephi does not have the rich liver-brown body coloration or a high dorsal collar or a very short radiolar web.

The pale flecked body surface of Sabellastarte sanctijosephi is similar to that of S. pectoralis and S. samoensis , but both species differ in having higher dorsal collar margins. Sabellastarte pectoralis also differs, in having the distal part of each radiole with paired ridges, and S. samoensis in having thick-shafted abdominal inferior chaetae and uncini with shorter shafts.

The coloration of some populations of Sabellastarte sanctijosephi with pale unbanded crowns, each radiole bearing a brown line along the outside of the rachis, is like that described by Haswell (1884) for Sabellastarte australiensis and observed in material from Arrawara (P.K.J.). Sabellastarte australiensis and S. japonica differ from S. sanctijosephi in having interdigitating radioles and bispiral crowns which are short, particularly considering their larger size.

Sabellastarte sanctijosephi may be confined to the Red Sea.

‘Sabellastarte’ inquirenda Sabella fusca Grube, 1869 .

Grube’s Red Sea material cannot be found in either the Berlin (ZMB) or Wroclaw (MPW) museums, where his other material is held. McIntosh’s (1885) record from Sydney is (as he suspected) Sabellastate australiensis , but his material from Sri Lanka (NHML 88.4.13.13) is S. spectabilis . Gravier’s (1906) figures (243–245) are of Bispira porifera (Grube, 1878) and Japanese records are also B. porifera (Nishi et al., 2000) . Rullier and Amoureux’s (1979) record off Rio de Janeiro (MNHMP A885 as Sabella ) is not a Sabellastarte as companion chaetae are present. The tube is covered with coarse shell and grit particles as are often found in Megalomma , but the distal half of the dorsal crown radioles are missing (some species of that genus have eyes only on dorsal radioles), so the species cannot be confirmed as being in Megalomma . Furthermore, the cuticle bearing the parapodia is mostly loose so the arrangement of abdominal chaetae cannot be determined. Gibbs’ (1971) record of Sabella fusca is actually Demonax aberrans (Augener, 1926, see Knight-Jones and Perkins, 1998).

Sabella grossa Baird, 1865 , from St Helena Island (NHML ZB 1972: 83 Type) is a large species (14 mm wide) with a bispiral crown of five whorls each side, but all radioles have been removed, perhaps by Baird, to show the five whorl bispirality of the crown base. The base of each radiole is present showing that they were not interdigitated, and the ventral sacs are rounded and external to the crown base, so it cannot be a species of Sabellastarte . Furthermore, it cannot be synonymous with the other large bispiral species Sabella longa (Kinberg, 1867, since put into Sabellastarte by Hartman, 1959: 561, and later into Pseudobranchiomma by Knight- Jones, 1994), because the dorsal collar margins of S. grossa are fused to the sides of the midline faecal groove. The chaetae within each abdominal fascicle (about 60, all similar and curved antero-dorsally) are laterally compressed to form an oval with a transverse axis and their arrangement (often generically indicative) is difficult to determine.

Species removed from Sabellastarte

Sabella assimilis McIntosh (1885) referred to Sabellastarte by Hartman (1959: 566) is a species of Perkinsiana Knight-Jones (1983) ; Sabellastarte arctica Ditlevsen (1937) is a species of Branchiomma Kölliker (Knight-Jones, 1994) histolysized except for a few stylodes, as is Sabellastarte indica var. oculata Rullier, 1964 (MNHN A391), for again the stylodes were overlooked. Sabella longa Kinberg (1867) and Dasychone odhneri Fauvel (1921) were wrongly synonymized by Hartman (1959) and Day (1967), as were Sabella zebuensis McIntosh (1885) and S. bocki Johansson (1922) , and the senior synonyms were put into Sabellastarte (Hartman, 1959) . All four species, however, belong to Pseudobranchiomma Jones, 1962 (Knight-Jones, 1994). Likewise Laonome punctata Treadwell, 1906 , once synonymized with Sabellastarte indica (Hartman, 1959: 548) , should be Pseudobranchiomma punctata (new combination), in spite of having a collar fused to the sides of the midline groove and lacking serrations on the outer surface of the radioles (Nogueira and Knight-Jones, in press). These species cannot be regarded as Sabellastarte because the ventral sacs are rounded, closely aligned and positioned outside the crown base (as in figure 1B).