Botrylloides israeliense, Brunetti, Riccardo, 2009

Brunetti, Riccardo, 2009, Botryllid species (Tunicata, Ascidiacea) from the Mediterranean coast of Israel, with some considerations on the systematics of Botryllinae *, Zootaxa 2289, pp. 18-32 : 22-30

Botrylloides israeliense n. sp.

Material examined: 7 colonies from the Bay of Àkko (May 1991), 2 colonies (? 2004), and 17 colonies (July, August, November 2007) from Mikhmoret.

PLATE 1. Botrylloides leachii . a, zooid, ventral view with eggs and testes: note masses of pigmented cells on each side of endostyle; b, zooid, dorsal view, showing larvae in incubatory pouches; c, oral siphon and dorsal lamina (internal appearance), some details enlarged; d, parietal view of gut loop; e1, e2, parietal and mesial sides of stomach; e3, e4, same from zooids of Adriatic colonies. (Scales: a, b, 1 mm; c, d, e1 – e4, 500 μm).

PLATE 2. Botrylloides anceps . a, zooid, dorsal view: note spots on each side of oral siphon; b, zooid, lateral view; detail of neural complex, enlarged; c, oral siphon and dorsal lamina (internal appearance), detail of neural complex, enlarged; d, dorsal lip of atrial opening; e1, e2, edge of oral siphon, relaxed and contracted, respectively; f, parietal view of gut loop, anal opening contracted and relaxed (detail); g1, g2, parietal and mesial sides of stomach of Israeli colonies; g3, g4, same from zooids of young colonies. (Scales: a, b, 1 mm; c, 250 μm; d, f, 100 μm; e1, e2, 50 μm; g1–4, 150 μm Symbols: cbw, connection to body wall; cf, ciliated funnel; cg, cerebral ganglion; dl, dorsal lamina; ng, neural gland; pg, pyloric gland).

PLATE 3. Botrylloides israeliense n. sp. a, zooid, left side; b, zooid (part of right branchial wall), note masses of pigmented cells on ventral sector of transversal vessels; c, oral siphon and dorsal lamina (internal appearance); d1, d2, parietal and mesial sides of intestine. Botrylloides magnicoecum (typus) e1, e2, parietal and mesial sides of intestine. (Scales: a, 1 mm; b, c, 500 μm; d, e, 250 μm. Symbols: cbw, connection to body wall; ocs, cross-section of oesophagus).

PLATE 4. Diagram showing structure of system in Botryllus and Botrylloides . Botrylloides : a, zooid; b, structure of system (branchial sacs not shown); e, meandric system; f, elliptical system. Botryllus : c, zooid; d, structure of system; g, star-shaped and h, elliptical system. (Symbols: ao, atrial opening; as, atrial siphon; cc, cloacal cavity; co, cloacal opening; dt, dorsal tongue).

Holotype: the type material consists of one colony from Mikhmoret, Israel, deposited in the Museum of Natural History of Venice, Italy.

Description. The colonies are usually 1.5 mm thick. The colour, in fixed material, is yellow. The test is transparent. The zooids are inclined as regards the colony surface, they lack atrial siphons, and are arranged in “ leachii type ” systems. There are 8 tentacles, 4 larger and 4 smaller, the first almost equal in length, without pigmented blood cells in the vascular lacunae at their bases. The dorsal tubercle has a more or less oval opening (Plate 3, c). The branchial sac, is conical with 8 rows of stigmata (the 8th is often difficult to detect) and all rows are complete (Plate 3, a). The branchial formula at the middle of the branchial sac is usually: 4,3,3,4 DL 4,3,3,4. There are no masses of pigmented blood cells on either side of the endostyle, but these cells accumulate in the ventral part of the branchial transversal vessel, especially from the 2nd to the 5th (Plate 3, a, b). The dorsal lamina rises at the level of the 3th or 4th row of stigmata (Plate 3, c). The stomach is as large as it is long, usually with 8 folds, excluding the typhlosolis. The folds of the surface turned towards the branchial sac (numbers 3 to 6) are longer and their pyloric end is not closed (Plate 3, d2). Folds 7 and 8 are progressively shorter, and there is sometimes a rudimentary 9th one. A smooth area is found between the last fold and the typhlosolis. The latter does not extend over the pyloric end of the stomach. From half-way along it, a caecum rises, almost long as the stomach. It is posteriorly inclined 45° to the long axis of the stomach, and is directed into the intestinal loop with a moderately dilated tip (Plate 3, d1). The duct of the pyloric gland is connected to the caecum about half-way along. The intestinal loop is peculiar: the terminal part of the intestine, usually empty and ribbon-shaped, is not bent but folded towards the interior; this peculiarity is not found when the rectum is full of faeces. Two grooves are present along its anterior and posterior edges. The oesophagus also has four longitudinal folds, with a cross-section in the shape of a four-leaved clover (Plate 3, d1–2). Although this latter character may be an artificial shape due to fixation, I do not think it is, as it has been found in all zooids from various analysed colonies. Only observation of living zooids will be able to clarify this point. The anterior edge of the intestinal loop reaches the penultimate row of stigmata, and a smooth edged anus opens at the same level or immediately anteriorly. Gonads are not visible.

Remarks. The presence of a large pyloric caecum initially suggested that this species was B. magnicoecum , originally described from the Cape of Good Hope, South Africa, by Hartmeyer (1912) as a variety of B. niger Herdman, 1886 , and thought to be present also at Eilath by Pérès (1962). Specimens from the three larger oceans have been assigned several times to this species on the basis of the presence of a large pyloric caecum. However, descriptions by authors are conflicting. Fortunately, I was able to analyse Hartmeyer’s typus, kept at the Naturhistorisches Museum of Vienna (a report on this revision will be published in the Boll. Nus. Civ. St. nat . Venezia). Hartmeyer’s species clearly differs in several morphological characters from the one from Israel described above, but the appearance of the intestine is sufficient to discriminate the two species (Plate 3, d1, d2, e1, e2). Both have a globular stomach with 8 (but sometimes 9) folds in israeliense and 9 in magnicoecum . Some folds, those on the internal surface, are opened in israeliense , whereas they are always posteriorly closed in magnicoecum . The typhlosolis is limited to the stomach surface in israeliense , but extends along the first tract of the intestinal loop in magnicoecum . In addition, in israeliense , the pyloric coecum is club-shaped, not curved, rising from half-way along the typholosis and not longer than the length of the stomach, whereas in magnicoecum it is finger-like, curved, rising from the posterior stomachal tract of the typhlosolis, and longer than the stomach. Lastly, in israeliense , the anus is smooth-edged and opens at the level of the anterior edge of the intestinal loop, whereas in magnicoecum it has two very evident lips and opens one or two rows of stigmata forward of the anterior edge of the intestinal loop.

The main characteristics of the three species described above are listed in the following table.

B. leachii B. anceps B. israeliense Test surface smooth sandy smooth 1. Botryllus - Botrylloides : validity of the genera

Savigny (1816, p. 198) first divided the genus Botryllus in two groups:one characterised by zooids cylindrical with branchial and atrial apertures close together, including Botryllus leachii ; and the other characterised by zooids ovoids with distant apertures, including Botryllus schlosseri . Milne-Edwards (1841) recognised the value of this distinction and raised the two groups to generic rank by founding the new genus Botrylloides . Herdman (1886) accepted Milne-Edwards’ point of view, but stressing the appearance of the systems (circular or elongate) was probably the unintentional cause of the misrepresentation of later authors who distinguished the genera exclusively on the basis of the external appearance of the systems, disregarding their internal structure. However, all colonies of Botryllinae have originally circular systems, so that only the internal structure of the system is important. Michaelsen (in Hartmeyer & Michaelsen, 1928), not considering the latter as a valid character (Die Art der Systembildung ist bierfür, wie ich jetzt glaube erkannt zu haben, ganz unbrauchbar. p. 321), gave a new definition of the two genera, based on the mutual position of the gonads: ovary anterior or dorsal to testicle, embryos developing in the peribranchial cavity: Botryllus ; ovary posterior to testicle, embryos developing in a broad pouch: Botrylloides . I recognise the validity of the two genera as defined by Milne-Edwards (1841, p. 300), whose very clear description was probably not completely understood. Briefly, the differences between the two genera consist of the presence or absence of an atrial siphon and in the consequent structure of the cloaca. In Botryllus zooids, the atrial opening, the anterior rim of which extends as a small dorsal tongue, is located at the distal end of a conical siphon (Plate 4, c). The cloaca is a small well, the wall of which is formed by the lateral fusion of the dorsal tongues of the atrial openings. All the zooids of a system are involved in the construction of the cloaca (Plate 4, d, g, h). This means that only a limited number of zooids can be arranged in a system. Systems with up to 8 zooids are usually typically star-shaped; when a greater number of zooids have to be arranged, the system become elliptical, and the same happens to the shape of the cloacal opening (Plate 4, g, h). Of course, ovoidation of the system involves lengthening of the atrial siphons of the zooids arranged at the two poles of the ellipse. This clearly has a limit, so that, when the number of buds which must substitute the filtering zooids of a system exceeds 15–20, they are arranged in two or three new systems during the generational changes. In Botrylloides , there are no atrial siphons and the atrial opening is wide, usually with a broad dorsal lip extending from its anterior rim (Plate 4, a). The zooids approach each other side by side, usually in an almost vertical position, defining a wide cloacal canal in which the branchial sacs of the zooids face each other. The roof of the canal is formed of the dorsal lips fused together (Plate 4, b). From time to time, there are cloacal openings along the roof of the canal, with edges formed by the unjoined dorsal lips of a few zooids (Plate 4, b, e, f). Such a structure, defined by Berrill (1950) with the felicitous expression “ladder system”, permits the formation of meandering systems of an unlimited number of zooids (Plate 4, e). Circular or oval systems are also possible; of course, the latter are the only ones present in young colonies. However, even in these small systems, the structure of the cloaca is the same, and it is easy to see that only a few dorsal lips are involved in the formation of the cloacal opening (Plate 4, f; compare f with h). Observing the cloacal opening from above in Botryllus , we see the openings of the atrial siphons of all the zooids of the system whereas, in Botrylloides , we can see the branchial sac of a few zooids near the cloacal opening. This different structure in the colonial system of the two genera was clearly described by Milne-Edwards (1841, p. 300–301), and the subsequent confusion was due to the use by later authors of expressions such “circular, “elongate or “double-row” systems. For this reason, I prefer to call the two structures “ schlosseri type ” and “ leachii type ”, respectively.

2. Some considerations on taxonomic characters in Botryllinae

Generally, authors describing a botryllid species mainly focus on the aspect of the colony, the number of rows of stigmata, the shape of the intestine, and the number and position of the gonads. Other features are sometimes added: number of oral tentacles, presence and distribution of pigmented cells, and mode of development of larvae. The aspect of the colony, excluding its structure, assumed to be a generic character (see above), is not a valid specific discriminating feature, as it is considerably influenced by the environment. The greater or smaller crowding of systems and the number and position of cloacal openings in Botrylloides species may change within the same colony. As first observed by Bancroft (1903b), no systematic value may be given to pigmentation, which is extremely variable ( Sabbadin & Graziani, 1967) and may change in different parts of the same colony as a consequence of different light conditions, as already observed by Giard (1872, p. 549). The presence, abundance and shape of ampullae between systems and along the edges of colonies are also without any systematic value, as they are functions of the physiological condition of the animals ( Brunetti et al., 1980).

The number of rows of stigmata is a less variable character. When colonies of Botryllus schlosseri and Botrylloides leachii developing in aquaria are studied from metamorphosis to the adult stage, the number of stigmata rows is seen to increase in subsequent blastogenetic generations, until it reaches a value characteristic for the studied species. This number is usually quite constant in the population of a given area but may change according to locality, as blastogenetic development is a function of temperature. However, the range of variation is small. Therefore when differences of 5–10 or more units are found, we are probably dealing with a different species. Certainly, the shape of the intestine is a very valid diagnostic character. Unfortunately, this organ is often poorly described, so that the same description may be proper to different species. Gonads are important, but they are not always present in filtering zooids of sampled colonies, and the same happens for larvae. Gonadic primordia occur in buds, but they are usually absent in adults for two or three blastogenetic generations, or even more when temperature conditions are not optimal ( Brunetti, 1974). During the generational change, the testes are reabsorbed, whereas oocytes are transported by the circulation to various parts of the colony, as hypothesised by Pizon (1893) and Izzard (1968) and experimentally demonstrated by Sabbadin et al. (1970). Their position in buds may also be different from that in adults. It is known that buds do not contract during fixation, but their morphology does not always correspond to that of adults. Organogenesis in buds is rapidly completed during the last hours before or during the generational change. So the probability of finding buds with definitive morphology is low: the number of rows of stigmata is usually lower, the tentacles are only outlined, and the intestine has not reached its definitive shape. As the development of branchia follows an antero-posterior gradient, whether the second row of stigmata is complete or not can only be clearly detected in buds. The number of tentacles may change within the same colony, because those of low order may not be completely developed but, as I believe I have demonstrated here, the shape and arrangement of those of the first order may be a valid diagnostic character.

The presence of masses of pigmented cells on both sides of the endostyle has been reported several times. Observations of living colonies of B. schlosseri and B. leachii explain their origin. During the generational change, a high number of large haematic cells (macrophages), which digest the tissues of the regressing zooids, are present in the circulation. When the new system is formed, cells accumulate on both sides of the endostyle, forming masses usually corresponding to the number of rows of stigmata. These masses persist for several days, and disappear only a few hours before the new generational change. A film of the microscopy of living colonies of B. schlosseri by Burighel (who kindly showed it to me) proves that, althogh single cells are removed from the mass by the circulation, their fate is still unknown. I cannot present ascertain whether the mass of macrophages at the base of the tentacles described above have the same origin and fate.

3. Sample area

The sample area belongs to the Mediterranean Levantine basin, characterised by oligotrophic waters with high temperature and salinity. The area is in communication with the Red Sea through the Suez Canal; in addition, in the last few decades, the Bitter Lakes have gradually equated their salinity to that of the Red Sea, and the construction of the Aswan High Dam across the Nile (in the 1960s) has reduced the eutrophic freshwater inflow from that river, so that its conditions have become even more like those of the Red Sea. So Lessepsian migrations are increasing ( Galil & Zenetos, 2002), and this fact may explain the presence in this area of the Indo-Pacific species Botrylloides anceps . The new species Botrylloides israeliense may also have the same origin. Instead, the extreme hydrological conditions make these waters very unfavourable for other Mediterranean species. The sample area probably represents a transition between the Mediterranean and Red Sea ascidian populations and, as Lessepsian species may be present, it is hoped that more extensive study can be made.