Redescription and systematic status of the Antarctic genus Abietinella Levinsen, 1913 (Lafoeidae, Hydrozoa, Cnidaria)

The Antarctic–Patagonian genus Abietinella, comprising two known species, Abietinella operculata (Jäderholm, 1903) and Abietinella grandis (Vanhöffen, 1910), is reviewed. The holotype of Abietinella operculata is fully redescribed, including morphometry and cnidome, unknown up to now. Its distinctive characters are the growth habit, hydrothecal shape and, most important, the presence of a dish‐shaped operculum attached to the adcauline side of the hydrothecal aperture. We corroborate its conspecificity with A. grandis and, therefore, the monotypic condition of Abietinella.


Introduction
The genus Abietinella was proposed by Levinsen (1913, p 294) to accommodate the only two known species, originally described as Zygophylax but bearing an operculum, namely Zygophylax operculata Jäderholm, 1903 andZygophylax grandis Vanhö ffen, 1910. The presence of an operculum and a caecum at the abcauline side of the hydranth led Levinsen to consider the genus in the family Sertulariidae, although subsequent authors had another opinion (e.g. Blanco 1968Blanco , 1976Vervoort 1972;Stepan'yants 1979;Rees and Vervoort 1987;El Beshbeeshy 1991;Peñ a and García-Carrascosa 1993;Blanco et al. 1994). Recent authors, concerned with the importance of the mixed features of the group, considered the genus as having ''intermediate characteristics between the families Lafoeidae and Sertulariidae'' García-Carrascosa 1993, p 1009).

Description of holotype
Stem erect, ca 45 mm high (SNHM: basal fragment 9 mm, medium 20 mm, mediumapical 9 mm, apical 7 mm; BMNH: basal fragment of stem 6 mm long with three hydrocladia, distal 9 mm long, and two branched hydrocladia 8 and 9 mm long), pinnate, polysiphonic, composed of main tube, divided into internodes, and a few secondary tubes involving main tube up to its distal part; branches arising from main stem tube; basal polysiphonic part of stem ca 0.93 mm, distal part 0.33 mm in diameter. No hydrorhiza present. Stem with branches up to third order; pedicels arising from main cauline tube and branches of first, second and third orders. First-order branches (ca 30) up to 30 mm long, 0.15-0.20 mm (0.16¡0.02, n512) in diameter, monosiphonic, occasionally the largest polysiphonic, planar, alternate; branches of same side arising 1.92-2.25 mm (2.03¡0.09, n514) apart from each other. Secondary and third-order branches, and hydrothecae, in the same plane as primary branches. Main cauline tube divided into internodes bearing apophyses (visible only at distal part of stem); cauline apophyses generally bearing alternately arranged branches and hydrothecae, although median-distal region sometimes with series of up to five hydrothecae in a row; distal stem bearing only hydrothecae. Stem apparently without nematothecae. Accessory tubes of polysiphonic part not divided into internodes and without apophyses, but with irregularly distributed nematothecae. Lateral branches arising at angles of 45-60u in relation to long axis of stem, with an axillary pedicellated hydrotheca without nematothecae at origin. Apophysis separated from branch by a well-marked septum. Branches not divided into internodes, with apophyses bearing lowerorder branches or pedicels. Pedicels short, continuous to apophyses, 0.11-0.32 mm (0.18¡0.07, n510) long at abcauline side, 0.05-0.22 mm (0.10¡0.05, n510) long at adcauline side, ending at diaphragm of hydrothecae. One pair of nematothecae at adcauline side of apophyses, one on each side of hydrotheca, rod-shaped with circular distal aperture, 0.20-0.23 mm (0.22¡0.01, n58) long, 0.04-0.06 mm (0.05¡0.01, n58) in diameter at aperture.

Additional data
Peñ a and García-Carrascosa (1993, p 1005) inferred the ontogeny of the fasciculation of A. operculata as: ''Main stem and lower-order branches are polysiphonic. However, A. operculata begins to grow by forming an erect, monosiphonic stem that emerges from a stolonal hydrorhiza, and carries hydrothecae set in the same plane. As the colonies develop, the stem is covered by secondary tubes that finally conceal the whole stem and its hydrothecae, in the basal area of the colony. The covered area decreases to the distal portion until it disappears close to the apex of the colony, where the stem is visible''. The authors indicated the presence of nematothecae also on the hydrorhizal stolons. Finally, the description of the gonosome by Peñ a and García-Carrascosa (1993, p 1009) can be synthesized as: coppiniae set along stem and main branches, fusiform, 9-17 mm long, 3-4 mm in maximum diameter, gonothecae closely aggregated around stem, elongate, flask-shaped gradually increasing in diameter from base to two-thirds of height, then hood-like apically with large lateral opening, height 1.08-1.12 mm, maximum diameter 0.40-0.43 mm; surface view of gonothecae with wider parts fused though keeping their individual perisarc delimitation, distal parts free with no preferential orientation for aperture; protective tubes with nematothecae arising amongst gonothecae of coppinia.

Remarks
There is a general lack of information in the literature concerning the cnidome of lafoeids (or Leptothecata in general), although it is known as a rule that the cnidome of Leptothecata is less diversified than that of Anthoathecata, for instance (see Bouillon 1985). The cnidome of A. operculata has not previously been studied, even though comparisons with the nematocysts of other leptothecates would be highly desirable. The nematocysts of the holotype of Abietinella operculata are of the same type (microbasic mastigophores), morphology (elongated oval) and equivalent dimensions (2263 mm cf. Boero 1980, Figure 7) as those of the cnidome of the Lafoeidae Hebellinae Hebella parasitica (Ciamician, 1880) (Boero 1980, 136, Figure 7), though Itô and Inoue (1962, p 449, Figures 78, 79) reported quite different dimensions for the same species (6.0-6.461.9-2.1 mm). However, we also studied nematocysts from a better preserved non-type specimen of A. operculata (USNM, Sta. 575/052, from Saunders Island) and the holotype of Abietinella grandis (ZMB Cni 958*, from Eastern Antarctic), and found three different size classes of nematocysts (Tables II, III; see description below), apparently belonging to two different types: large and medium ?macrobasic mastigophores and small microbasic euryteles. Nematocysts of the type specimen of A. operculata are intermediate in size between the large and medium-sized nematocysts of the specimens from Saunders Islands and Eastern Antarctic and, once all determinations are dubious, it is very plausible they are correspondent. We possibly overlooked two size-classes of nematocysts in the type because it is not well preserved. Therefore, we conclude that the cnidome of Abietinella is variable concerning nematocyst dimensions, and different from that of known lafoeids Hebellinae. The cnidome of Abietinella is also different in some aspects from that of the sertulariid genera used by Levinsen (1913) to justify the placement of Abietinella among the Sertulariidae. In general, the species of Sertulariidae do not have three size classes of nematocysts. Diphasia tropica Nutting, 1904 has microbasic mastigophores in two dimensions, 7.5-8.062.5-3.0 mm and 5.0-5.561.5-2.0 mm (see Migotto 1996). The cnidome of Abietinaria, on the other hand, is also constituted of microbasic mastigophores (e.g. 16.9-19.866.2-7.1 mm and 5.5-5.961.8-2.0 mm for Abietinaria costata Nutting, 1901) Migotto 1996]. In any instance, the cnidome of A. operculata is somewhat different from both Lafoeidae (in this case Hebellinae) and Sertulariidae.
Zygophylax grandis (5Abietinella grandis) was described by Vanhö ffen (1910) from material from the Gauss Station (65u219S, 86u069E, Davis Sea, Antarctica, 385 m, collected during the German South Polar Expedition), who considered the possibility that the five forms referred to Zygophylax at that time were conspecific. Vanhö ffen (1910) did not report nematothecae on the main stem tube, and acknowledged hydrothecae and nematothecae are suppressed on the internal tubes. The differences remarked by Vanhö ffen (1910) to justify his new species were the middle branches not lying on the same plane of the whole colony, general dimensions and proportions of hydrothecae and nematothecae. Although the holotype of Z. grandis was supposed to be lost (cf. Rees and Vervoort 1987) and the original description was apparently based on infertile material, the species was considered conspecific with A. operculata by Stepan'yants (1979; arguing that the size of hydrothecae may be, indeed, variable) and Vervoort (1972, p 82), Rees and Vervoort (1987, p 46) and El Beshbeeshy (1991, p 92) concurred.
We had the opportunity to study the holotype of A. grandis and may provide a re-description of the species, as follows.

Description of holotype of A. grandis
Stem erect, ca 105 mm high, basally and distally broken, pinnate, completely polysiphonic. Stem composed of main tube and a few accessory tubes involving main tube up to its distal end; branches arising from main stem tube; basal polysiphonic part of stem ca 1 mm in diameter, distal polysiphonic part ca 0.4 mm. Stem basally broken; no hydrorhiza present. Stem with branches up to second order. Monosiphonic branches up to 14 mm long. Only four primary branches, at 55, 75, 84, and 105 mm from stem base, giving rise to secondary branches; all forked primary branches, except the last one, polysiphonic and up to 15 mm long. Polysiphonic, primary branches arising at a plane different from that formed by remaining first-order branches and hydrothecae; the first two at almost right angle.
Hydrothecae alternately arranged in one plane, but sometimes in two planes making an obtuse angle. Hydrothecae arising at ca 45u in relation to longitudinal axis of branch. Hydrotheca tubular, adcauline wall sinuous, widely convex basally and concave at distal part; abcauline wall usually straight, but occasionally slightly curved as well. Hydrothecae 0.656-0.720 mm (0.683¡0.017, n510) long (from diaphragm to margin at abcauline side), with fine striae on outer surface. Base of hydrotheca 0.112-0.144 mm (0.132¡0.01, n510) in diameter at diaphragm level; maximal diameter of 0.224-0.248 mm (0.235¡0.009, n510); diameter at aperture 0.208-0.248 mm (0.229¡0.012, n510). Hydrothecal wall slightly thickened basally. Diaphragm thick, circular, usually slightly directed downwards in relation to hydrothecal long axis. Hydrothecal aperture circular, oblique in relation to hydrothecal long axis, rim even and slightly flared, and usually with a few short renovations. Hydrotheca closed by a disc-shaped operculum, attached to adcauline wall and projecting into the hydrotheca at 30-45u; operculum frequently lost. Hydranths tubular, basal region constricted by diaphragm, median region often with well-developed abcauline triangular caecum, apical region with one whorl of ca 12 tentacles, hypostome conical.
Because of the polysiphonic development it is not possible to determine whether or not the main cauline tube is divided into internodes. Apparently, the stem is old, being almost deprived of hydrothecae, except the distalmost 30 mm, where it is possible to see the hydrothecal and branch arrangement, alternate and in one plane. However, at distal part of the stem it is also possible to find a few hydrothecae arising from an accessory tube, probably because the original stem was broken. In fact, at the end of the stem there are two primary branches arising at the same level, apparently one from the original main tube and another from an accessory tube. One of them is unforked, but the other is branched and gives rise to two contiguous second-order branches alternately arranged in one plane. This forked, primary branch is not divided into internodes.
Peñ a and García-Carrascosa (1993, p 1010) also noted the resemblance of A. operculata and A. grandis, but admitted the existence of some differences: ''in A. grandis, nematothecae are very long and narrow, with a small opening (ratio 6:1), while in A. operculata they are short and wide with a large opening (ratio 1.5-3:1)'', the hydrothecae of A. grandis ''are bigger, and the branches can emerge in more than one plane''. However, Argentina (Blanco 1968(Blanco , 1976 Argentina (Vervoort 1972) Argentine shelf (Stepan'yants 1979)  a comparison with the holotype and other material of A. operculata (namely Blanco 1968, 1976, Vervoort 1972Stepan'yants 1979;El Beshbeeshy 1991;Peñ a and García Carrascosa 1993) indicates that A. grandis (Vanhö ffen, 1910) falls within the size range of Abietinella operculata, except for the length of the hydrothecae and nematothecae. Correlation between morphometric data and the known distribution of the species, restricted to Antarctic and sub-Antarctic waters, may suggest that hydrothecae from high Antarctic areas are larger (0.72-0.75 mm in length and 0.22-0.24 mm in diameter at aperture for material from the Wedell Sea; see Peñ a Cantero et al. forthcoming). Therefore, we conclude that the hydrothecal and nematothecal lengths are variable and do not constitute enough evidence to support two different species.
Considering the previous descriptions of A. operculata, one can notice that the largest variation in the group concerns morphometry, but a gradual series of morphometric data can be observed (see above and Tables I, III, IV). The few morphological discrepancies amongst the different colonies of A. operculata were found in material recorded from several stations in the Patagonian region by El Beshbeeshy (1991, p 90), differing from the holotype in the abcauline attachment of the operculum and in the monosiphonic stem, though this last difference is possibly due to the less-developed or younger colonies.
The systematic position of A. operculata has been in dispute since its discovery. The species was primarily referred to the genus Zygophylax (Lafoeidae after Quelch, 1885, in his original description of the genus) by Jäderholm (1903), because of similarities concerning the morphology of the colony, hydrothecae, nematothecae, and abcauline blind sac; the same classification was followed by Billard (1905Billard ( , 1906, Jäderholm (1905), Clarke (1907), and Vanhö ffen (1910). Levinsen (1913) pondered on the existence of similarities (namely presence of operculum, hydrothecal and diaphragm shape) between A. operculata and the species of the genera Abietinaria Kirchenpauer, 1884 and Diphasia L. Agassiz, 1862, both considered to belong to the family Sertulariidae. This proposal of the sertulariid affinities of Abietinella was followed by Naumov and Stepan'yants (1962), disregarding the possibility of the existence of an operculum and caecum in any family but Sertulariidae. Agreement about the validity of the genus Abietinella, although placed among the Lafoeidae, was given by many authors (e.g. Blanco 1968Blanco , 1976Vervoort 1972;Stepan'yants 1979;Rees and Vervoort 1987;El Beshbeeshy 1991;Peñ a and García-Carrascosa 1993;Blanco et al. 1994). On the contrary, Salvini-Plawen (1972), although regarding the species among the lafoeids, referred it to Cryptolaria operculata, a position neither justified nor followed by anybody else. Besides, the binomen Cryptolaria operculata had already used by Nutting (1905, p 947-948) to describe a hydroid from Hawaiian waters, presently named Stegolaria operculata (Nutting, 1905) (e.g. cf. Stechow 1913a1913b, p 137;Edwards 1973, p 594;Hirohito 1995, p 94).
The doubts concerning the taxonomic position of Abietinella operculata began to be clarified when Stepan'yants (1979) firstly observed specimens with coppinia, however not describing it in detail. Later, Rees and Vervoort (1987, p 46) also regarded the gonophores of the species as typical coppiniae, found on main stem and branches. However, a detailed account of the coppinia was only given by Peñ a and García-Carrascosa (1993). These data corroborated the position of A. operculata within the family Lafoeidae.
Manuel García-Carrascosa (Universidad de Valencia, Spain) for accommodating us in his lab for part of the studies, and to an anonymous referee for reviewing the text. We specially thank Dr Wim Vervoort (Nationaal Natuurhistorisch Museum, Leiden, The Netherlands) for his help with the bibliography, careful revision of the text, and inestimable collaboration in finding the type of A. grandis, that significantly improved the manuscript. The project had financial support from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 2001/02626-7 and2001/10677-0), from the National Museum of Natural History, Department of Invertebrate Zoology, United States Antarctic Program funded by the National Science Foundation (OPP-9509761), and from the Ramó n y Cajal Program co-sponsored by the Ministerio de Ciencia y Tecnología and the Universidad de Valencia of Spain. A.C.M. and A.E.M. also have financial support from the Conselho Nacional de Desenvolvimento Científico e Tecnoló gico (CNPq;300271/2001-8 and300194/1994-3, respectively).