Protula Risso, 1826

Ben-Eliahu, M. Nechama & Ten Hove, Harry A., 2011, Serpulidae (Annelida: Polychaeta) from the Suez Canal- From a Lessepsian Migration Perspective (a Monograph) 2848, Zootaxa 2848 (1), pp. 1-147 : 45-49

Genus Protula Risso, 1826

Table 1

For a generic diagnosis, see ten Hove & Kupriyanova (2009: 81–83).

Preliminary remarks. In their recent review of the taxonomy of serpulid genera, ten Hove & Kupriyanova (2009: 81) classified Protula as the most problematic serpulid taxon. That is because the phylogenetic basis for this genus is ill-defined and based on negative characters, such as lack of an operculum, lack of special collar chaetae and lack of any characteristic ornamentation of the tubes (mostly). Furthermore, characters for species distinction are scarce and poorly understood (ten Hove 1984: 187; 1994: 109). Not only were the initial descriptions scanty ( Table 1), but nearly all were based on only one or, at the most, a few specimens (ten Hove & Pantus 1985: 421), so that variability in taxonomic characters could not be taken into consideration. Many taxa have been described on at least partly-presumed differences in chaetation (ten Hove 1994: 109), such as, the characters, presence and location of Apomatus chaetae, that figure so importantly throughout the Protula literature, e.g., Uchida (1978) and in Table 1.

To evaluate the extent of variability of diagnostic characters in Protula, ten Hove & Pantus (1985) carried out a study of live Mediterranean individuals considered to be Protula tubularia ( Montagu, 1803) . Within the same population, they found individuals that possessed Apomatus chaetae and others that lacked them (ten Hove & Pantus 1985: 425; ten Hove & Kupriyanova 2009: 81). Apomatus -chaetae, extremely transparent, are very difficult to discern within the thick bundles of limbate chaetae, so that while presence data can be accepted as valid, absence data might be evaluated as provisional, particularly when the data derive from examination of whole animals under the compound microscope. On the other hand, examining dissected parapodia under high magnification should enable observing Apomatus chaetae if present ( Fig. 22F), as well as examining whole specimens with the SEM, where transparency has no relevance. In the studied Mediterranean Protula tubularia population, the location of rows of thoracic uncini also showed variability, with uncinigerous tori typically present from chaetigers 2–7. However, uncini might also be completely lacking, and, particularly in larger specimens, there was an impression that uncini might disappear secondarily (ten Hove & Pantus 1985: 429). The variability in both these characters illustrates why inferences based on few specimens should be viewed with the greatest caution, including inferences from literature. The urgently needed revision of this genus must therefore be based both on a comparison of all available types, and of sufficient topotypical material assembled for each taxon to produce statistically evaluated morphological criteria. The proposed review should include modern molecular techniques, providing supplementary evidence as to whether morphologically similar (allopatric) populations belong to one or more taxa (e.g., Golani & Ritte 1999, Bucciarelli et al. 2002 and Bonhomme et al. 2003).

In our objective of dealing with the Suez Canal as a pathway and / or as a habitat, our first priority is elucidation of the source of the population inhabiting it. A first step entails assembling information on the taxa of the biogeographic regions adjoining the canal, and then excluding those taxa that do not correspond with the present material. (This approach should be carried out with other problematic taxa, as well, e.g., with Vermiliopsis sp. / spp. [see below]). Two Protula taxa from the Mediterranean must be taken into account: A larger taxon, Protula intestinum ( Lamarck, 1818) , type locality, “European seas”, 8–12 cm in length / 8–12 mm in width, and a smaller one, Protula tubularia ( Montagu, 1803) , type locality, England, 2–5 cm in length, 3–8 mm in width ( Fauvel 1927: 382–384). In surveying the tropical Indo-West-Pacific Protula literature, the records can roughly be brought together into two taxa: (a) a larger taxon with branchial radioles arranged in a spire of 3–8 whorls, with a tube-diameter of up to 2 cm (ten Hove 1994: 109; Smith 1985: 92–97, 433, fig. 9e), named Protula bispiralis (Savigny, 1822) , type locality, “les mers des Indes” (it was collected by F. Péron, thus, probably Australia [ten Hove 1994: 109; Weinberg 1996]) and, (b) a smaller one, Protula palliata , type locality Sri Lanka, described from a single specimen and characterised by a semicircular arrangement of the branchial radioles, at most in a 3/4 whorl, with a tube diameter of a few millimetres ( Table 1). In Willey’s type specimen, the body measured 19 mm in length (without branchial lobes), 3 mm in width; 30–40 radioles per lobe. Willey named it for its most exceptional character, the collar, with very long lateral lobes “rolled on themselves” and “probably able to follow the branchial spire to its termination. …” and “ventral border was slightly concave” ( Willey 1905: 316; Table 1). Other characters: Thorax with 7 chaetigers with limbate chaetae, and with Apomatus chaetae and uncini from the 4 th chaetiger; thoracic uncini, P+~20 teeth; Willey only illustrated an Apomatus chaeta, a sickle-shaped abdominal chaeta, and a profile view of the peg of a thoracic uncinus ( Willey 1905, pl. 7, figs 183–185). The taxon, Protula palliata , as recorded in the literature, is possibly a complex of species (ten Hove 1994: 109).

Twenty years before Protula palliata was described, McIntosh (1885: 511) published an Indo-West- Pacific Protula taxon, P. arafurensis , collected in the Arafura Sea by the Challenger Expedition. Unfortunately, the description was based on a single specimen lacking its branchial crown. As the species cannot be identified with certainty (ten Hove 1994; Table 1), P. arafurensis can only be regarded as a " species inquirendam ", one that cannot even be taken into consideration for species identification ( Table 1). The single later record, Treadwell’s (1906) Protula ? arafurensis , is erroneous, the material was re-examined by one of us (HAtH) and belongs to the Vermiliopsis infundibulum / glandigera –complex. Protula procera Ehlers, 1918 , type locality Aru Islands, Indonesia, differs from P. palliata in its short collar, larger size, larger number of radioles in a single lobe (56), and in a pectinate (rather than circular) arrangement of the radioles; the abdominal chaetae appear sickle-shaped ( Ehlers 1918 pl. 17, fig. 9). Protula anomala Day, 1955 , type locality South Africa, and also reported from E. Australia ( Smith 1985: 87–92), also can be excluded since it is clearly differentiated by its compound radiolar eyes and retrogeniculate abdominal chaetae (for a definition of retrogeniculate, see ten Hove & Kupriyanova 2009: 26).

The present Suez Canal specimens from the Bitter Lakes are small, with a tube diameter of <4 mm ( Figs 21A, D ), see below, and the branchiae are arranged in two semi-circles ( Figs 17A, B and 19A–C ). Thus, we can confidently exclude the Indo-Pacific taxon, Protula bispiralis , based on its spiralled branchial structure and large size, as well as the Mediterranean taxon, Protula intestinum , on account of its large size. (We have studied material of both nominal taxa from areas more or less adjacent to the Suez Canal). With much greater caution, we need to approach the identification of the smaller taxa from the Indo-Pacific and the Mediterranean that appear to be rather similar morphologically, as the mixed synonymies below suggest. Here the problem of allocation to species is exacerbated by Fauvel’s having included the Persian (Arabian) Gulf and the Indian Ocean as part of the range of the Mediterranean species, Protula tubularia , in his “Faune de France,” with a citation of the Indo-Pacific taxon, P. palliata , as a questionable synonym of P. tubularia ( Fauvel 1927: 382–383) . In his earlier Persian Gulf paper, Fauvel (1911: 433), the Protula included was P. palliata , and there Fauvel questioned whether P. tubularia was its synonym. However, in the report of the “ Siboga” Expedition to the Malay Archipelago, Mesnil & Fauvel (1939: 35) listed a Protula species as P. tubularia , with P. palliata as its synonym, and in that paper the authors noted that, “as concerns P. tubularia and P. palliata , neither the tubes, nor the worms showed differences sufficient to distinguish between them.” Fauvel’s (1953) “Fauna of India ” again referred to the nominal species, Protula tubularia , and cited Protula palliata as its synonym. The illustrations were copied from the “Faune de France ” (1927 fig. 257). We will never know whether, in arriving at this synonymy, Fauvel took into consideration the collar structure described by Willey, or whether he assumed the character to be of no particular importance, aberrant, or just described in an exaggerated way. In the later Indo-Pacific literature, Wesenberg-Lund (1949) listed Protula palliata and remarked on its very large thoracic membranes; Dew (1959) also listed P. palliata (citing Fauvel’s 1911 publication); as did Pillai (1960: 5–7).

Fauvel’s taxonomic (biogeographic) philosophy, especially on the occurrence of wide-spread “cosmopolitan” species, impacted not only on his own work but on an entire generation of polychaete researchers as well as on the present generation of non-specialists. In the following section on the genus Salmacina , we quote from Fauvel’s remarks on his predisposition to regard similar taxa from different biogeographic regions as belonging to the same taxon ( Fauvel 1933b: 144). In mitigation, Fauvel, who died in 1958 (age 92), seven years before the Scanning Electron Microscope became commercially available ( Breton 2006), may have been unaware that he lacked an adequate tool for comparison of minute morphological structures. Moreover, in 1933, when he published the report on the Dollfus Expedition polychaetes, he would have been unfamiliar with the concepts of invertebrate sibling species, first introduced by E. Mayr in 1942 (see also Mayr 1963; Grassle & Grassle 1976), and of cryptic species (e.g., Bucciarelli et al. 2002; Agapow et al. 2004).

1)

It is not possible to be absolutely certain that the 5 records of Protula palliata all refer to the same taxon. — 2) McIntosh’s description of “ Protula arafurensis ” is too incomplete to use for identification; P. arafurensis should be regarded as a “ species inquirendam” (see p. 46, above).— 3) Although the details of McIntosh’s (1885, fig. 18) thoracic uncinus are not clear, the number of horizontal rows of teeth above the peg may come to 23/24 rows.

In order to minimize the potential for error due to synonymising similar taxa from different biogeographical provinces, we propose, in contradiction to the philosophy expressed in Fauvel (1933b), to adopt the most conservative biogeographical approach possible, treating Protula taxa from the Atlantic- Mediterranean and from the Indo-West-Pacific biogeographical provinces as separate species unless / until synonymy can be proved for them. It then follows that the name, Protula tubularia , should be reserved for the Atlantic-Mediterranean forms—as noted, the type locality is England ( Montagu 1803)—and P. tubularia -like taxa occurring in the Indo-West-Pacific region should provisionally be assigned to P. palliata or to P. cf. palliata including the two unidentified Protula taxa with un-spiralled branchiae from the Seychelles and Amirantes Islands (ten Hove 1994: 109). We intend to continue looking for robust characters to distinguish between these taxa, particularly as the Suez Canal joins both the Atlantic-Mediterranean and Indo-West- Pacific regions, in order to establish the provenance of the Suez Canal Protula species that colonized it.

Molecular techniques are not presently applicable to denatured formaldehyde-fixed specimens; thus for comparing the older specimens found in most museum collections, we can still only employ morphological techniques. We consider that some characters used in the past to distinguish between species of the genus Protula are not sufficiently robust to sufficiently discriminate between similar species. For example, the question of type of abdominal chaetae, whether sickle-shaped or geniculate, that Fauvel (1927: 382) used to discriminate between the two Protula species from the Mediterranean, may exist in reality, and may prove to be a workable character to differentiate between taxa. However, ten Hove & Pantus (1985, fig. 2i–k) showed that the differences between the character states “sickle” and “geniculate” are not absolute but gradual, e.g., form of the abdominal chaetae varied within a single bundle from straight to geniculate, also with more or less sickle-shaped ones, and suggested that this character should be reserved for populations rather than for single specimens. In this regard, note the variability of shapes in abdominal chaetae of Protula specimens collected from underneath the same small rock in the Little Bitter Lake that ranged from sickle-shaped to geniculate ( Figs 18G–J ). We have even observed these different forms in a single individual from the Little Bitter Lake. Fauvel himself noted a reversal of the contour of abdominal chaetae produced by exposure to different chemicals ( Fauvel 1927: 384, legend of fig. 130h, i), putting in question the differences in contour of abdominal chaetae as a robust taxonomic criterion. Differences in the collar structure also may not be a reliable criterion (ten Hove & Pantus 1985: 427). A character that may prove useful in fresh Protula material (ten Hove & Pantus 1985: 429–430), but not in preserved specimens, is the presence / absence of radiolar and prostomial eyespots, because eyespots observed in freshly described specimens may fade in preservative ( Ben-Eliahu & Fiege 1996; Table 2), and their durability in preservative is insufficiently well known (e.g., paired crimson eyespots densely positioned on the rachides of Spirobranchus tetraceros radioles faded within three years of formalin-fixation and preservation in alcohol [ Ben-Eliahu et al. 2003]).

Presently, the best and most reliable morphological comparisons appear to be of the ultrastructure of the thoracic and abdominal uncini, using SEM micrography as a standard tool to determine, in frontal view, the number of rows of teeth and the number of teeth in the horizontal rows, the shape of the anterior tooth (in Protula , the “peg”), whether pointed, rounded, more or less swollen, truncated, “square-ended” or bilobed (e.g., Figs 18C, D, K, L ; Fig. 20A, P :?4:4:4:3:3:1:1:1:1:etc.; Table 3), while the light microscope should be used as well for squash preparations to provide a profile view ( Fig. 20B ).

Mediterranean material of the nominal Protula tubularia has been studied with SEM by ten Hove & Kupriyanova (2009) and again by the present authors. However, even with the SEM, due to the very close packing of the uncini in the torus, adjacent structures may be difficult to distinguish clearly. Thus, to better visualise the structure, uncini of Protula tubularia have been “cut” out of the photograph of the torus ( Fig. 18K [thoracic uncinus, cut from ten Hove & Kupriyanova (2009 fig. 39c), see detailed description in the legend]; and Fig. 18L —abdominal uncinus of specimen from the same sample) to compare with the abdominal uncini from specimens from the Little Bitter Lake ( Figs 18C, D ).

The following key presents some differences noted between these uncini:

1a. In the Suez Canal Protula taxon, both thoracic and abdominal uncini are of “saw-to-rasp-shaped” type, i.e., the teeth in the posterior apical (saw”)-part are in single file ( Figs 18C, D ); the peg tip is bilobed, with rounded lobes ( Figs 20D–G ); the thoracic uncini have ca. P+19 rows with ca. 3–?4 multiple rows, grading to?4 minute teeth in the row proximal to the peg ( Fig. 20A ) .

1b. In the Mediterranean Protula tubularia ( Figs 18K, L ), in both the thoracic and abdominal uncini, the teeth in the apical part form an angled double row, the uncini are thus “rasp”-shaped type in both apical (posterior) and proximal (anterior) parts and the tip of the peg appears more blunt (possibly abraded); moreover, the number of teeth is much greater; the thoracic uncini have ca. 21 apical angled double rows and ca. 17 multiple anterior rows grading to ca. 5–7 minute teeth in the rows proximal to the peg, with ca. 7 teeth in the proximal row, ca. 38 rows of minute teeth (counted from enlarged figs 39a, c in ten Hove & Kupriyanova (2009) and Fig. 18K ).

These differences in ultrastructure of the uncini point to the Mediterranean and Suez Canal Protula taxa as being different. Although recognizing that the ultrastructure may be variable and that the small number of specimens examined lacks a statistical basis of confidence, these differences presumably justify excluding the Mediterranean taxon, P. tubularia , as a founder population for the Bitter Lake specimens and support a Red Sea derivation for the population. That inference is bolstered by the fact that Beets’ Protula tubes were found on molluscs dredged from the floor of the Great Bitter Lake, thus, not connected directly to ship fouling, more so, as foulers of this genus have not yet been recorded from the Indo-Pacific (ten Hove 1994: 109). Another factor supporting the inference of a Red Sea derivation is the lecithotrophic nature of the Protula larva that dictates a short-term larval life and rapid settlement ( Kupriyanova et al. 2001; See Discussion, Section 6a). Taking these factors and the taxonomic approach discussed above into consideration, we designate the Protula species from the Suez Canal as P. cf. palliata , subject to confirmation by comparison with topotypical P. palliata material. (Meanwhile, efforts to obtain topotypical specimens have not yet proved successful. Moreover, the National Museum Colombo, where A. Willey had been a director, does not have any specimens [M. Goonatilake, NMC and Dr. T.G. Pillai, NHM, pers. comms.]).

Figures

FIGURE 18. Protula cf. palliata from the Little Bitter Lake. Chaetae and uncini (HUJ-Poly-1569, 1570, 4427, 4429, sample SBE1). A—Thoracic uncinus, saw-to-rasp-shaped (spec. 1569) with ca. 19 teeth in profile view (P+19), length 39 µm from apex to tip of peg, B—Abdominal uncinus from mid-body saw-to-rasp-shaped (spec. 4429) with ca. 24 rows of teeth in profile view, length 45.9 µm (pegs of additional uncini displaced when squashing the preparation seen in bottom left and right; left of arrow, part of uncinus inserted in the body wall, C, D—Frontal (edge) view of abdominal uncini “cut” from SEM micrographs of the tori, C—Spec. 4427, abdominal uncinus, with apex partially turned to the side, with ca. 24 teeth in profile view; from the apex, 18 teeth in single row, followed by ca. 2 rows with 2 teeth, then 2 rows of 3 teeth, and 2 rows of 4 teeth, arrow demarcates saw-part of the uncinus from rasp-shaped part; length, ~34 µm (magnification 6 K), D—Spec. 1570, dorsal-most uncinus with 24 teeth in profile view; from the apex, ca. 18 teeth in single file, followed by 2 rows of two teeth, then 2 row of 3 teeth and 2 rows of 4 finer teeth proximal to the bilobed, rounded peg; (arrow demarcates saw-part of the uncinus from rasp-shaped part); length ~28 µm (magnification 9 k), E—Spec. 1570, SEM micrograph of part of the torus, magnification 9 k; location of “D” in torus marked by arrow (another SEM of this torus, presumably most of the torus, magnified only 3 k, numbered at least 65 abdominal uncini); note that dorsal-most uncini are longer than the ventral uncini; 6th more ventral uncinus measures 21 µm, F—Spec. 1569, thoracic chaetae limbate (bordered) capillary chaetae only, Apomatus chaetae not observed, G–J—Abdominal chaetae from different specimens belonging to the same sample: G—Rather sickle-shaped with a somewhat prominent “heel” (spec. 4429), H— Somewhat concave proximally, less sickle-shaped than G (spec. 1569), I—Somewhat geniculate (spec. 4427), J– Both chaetae somewhat geniculate but also somewhat concave (spec. 4427 [1819]), K—Thoracic uncinus of Mediterranean Protula tubularia (Banyuls, France, ZMA.V.Pol. 3816), 1st torus (adapted from ten Hove & Kupriyanova, 2009, fig. 39c); from apex, 16–15 angled double rows of teeth, followed by ca. 3 rows of 3; 1 row of 4, 3 rows of 5, reaching ca. 7– 8 teeth in rows above bilobed peg, length of uncinus ~49.6 µm; arrow demarcates posterior-part of the uncinus from rasp-shaped part with more teeth in row, L—Anterior abdominal uncinus of Protula tubularia from same population as K, as in thoracic uncini the distal part comprises double row, ca 16 angled rows; proximal part, demarcated by arrow, comprises 9 rows with numbers of teeth increasing from 3 to ca. 7, ~27.3 µm (Table 2). For Suez Canal specimen codes, see Protula section, Hebrew University-Smithsonian Expeditions, sample SBE1).

FIGURE 19. Protula cf. palliata from the Great Bitter Lake (B&T-C-Pol 27- HUJ-Poly-1567), ca. 87 chaetigers, body ~27 mm length + ~5.6 mm branchia = total length ~32.6 mm; width, ~2.2 mm. A—Branchial lobe, inside view; arrow designates dorsal ovoid mouth palp; 19+?few more radioles, B—Enlargement of base of lobe to show the palp, C— Branchial lobe, outside view, D—Radiole; arrows designate white blotches; note the tip appears less filiform than that in Fig. 16A (?likely due to preservation), E—Low interradiolar membrane of branchial lobe at base of branchial lobe (arrows), F—Body, thorax of 7 chaetigers and abdomen of ca. 80 chaetigers (counted from photograph), 3/4 view dorsal side up; note dorsal pygidial glandular pad, G—Thoracic chaetae of 7th (last) chaetiger, Apomatus chaetae not present,. H, I—Abdominal chaetae, sickle-shaped: H—SEM micrograph, 2,400 x, I. Microscopic view with oil immersion. Scales: C, D—1 mm, F—5 mm; G—100 µm.

FIGURE 20. Protula cf. palliata from the Great Bitter Lake, uncini (B&T-C Pol 27-HUJ-Poly-1567). A—Part of torus of thoracic uncini from 7th chaetiger, ca. P+19 teeth in profile view, from apex: 14 teeth in single row, two rows of 3 teeth, 2–?3 rows of 4 teeth (left figure shows method of counting the teeth), B—Abdominal uncinus, profile view, left arrow shows part of uncinus inserted in body wall, right arrow designates section of uncini to the peg corresponding to Fig. D, numbering 23 teeth in profile view, length 46.7 µm, C—Two adjacent abdominal uncini from torus, left uncinus numbering 22 teeth in profile view, length 47.5 µm, D–F—Three views of enlarged peg, D—View taken under oil immersion, showing bilobed tip of peg, 4 teeth in row proximal to the peg; arrow shows part of uncinus corresponding to that designated by right arrow in Figs B, E—SEM micrograph showing frontal view of peg with 4 teeth in the two rows proximal to the peg (4,000 x magnification), peg appears abraded, F—SEM views of abdominal uncini showing markedly bilobed peg; magnification, 6,000 x, G—Part of torus of abdominal uncini from which F was taken, as indicated by arrow; the SEM micrograph of incomplete torus numbered 148 uncini (ends of torus not included on micrograph).

FIGURE 21. Protula cf. palliata tubes from the Bitter Lake with distinctive sculpturing—simple growth striations, but with the rim of the striation seemingly eroded with a somewhat scalloped appearance. A–C—Tube fragment of Protula sp. on Chama asperella from Beets’ Stn 15, 10.4 m (Beets 1953): A—Inner view of Chama valve bearing tube of 3.6 mm diameter, B—Lumen of same tube fragment, diameter ca. 4 mm, C—Enlargement of A, D—Tube fragments of Protula specimen from “Yellow Fleet” encrustation (B&T-C-Pol 27-HUJ-Poly-1567), legit H. Brattström and J.P. Taasen, 1975, diameter 4.8 mm. Scales, 5 mm.