|Home | About | Journals | Submit | Contact Us | Français|
The tintinnid ciliate Stenosemella pacifica Kofoid and Campbell, 1929 was occasionally recorded from the pelagial of temperate, subtropical, and tropical neritic waters. Since its cytological features were unknown, the species is redescribed from material collected in the pelagial of the Irish Sea, using live observation, protargol impregnation, and scanning electron microscopy. Furthermore, the species diagnosis is improved to include new characteristics, e.g. the somatic ciliary pattern comprising a ventral, dorsal, and posterior kinety as well as a right, left, and lateral ciliary field. The stomatogenesis of S. pacifica is typical for species with such a complex somatic ciliary pattern: the oral primordium develops hypoapokinetally posterior to the lateral ciliary field. The presence of windows in the lorica collar of Stenosemella ventricosa, the type of the genus, necessitates (i) an improved genus diagnosis, (ii) a synonymization of the genus Luminella Kofoid and Campbell, 1939, and (iii) a transfer of the Luminella species to the genus Stenosemella, including Luminella neocalifornica, which becomes Stenosemella neocalifornica nov. comb. Owing to the lack of a description, Stenosemella crateri is considered a nomen nudum.
THE classification of the about 1,200 tintinnid choreotrichid species is merely based on lorica morphology, as cytological features, including the ciliary pattern, are only known in 17 species (Agatha and Riedel-Lorjé 2006; Blatterer and Foissner 1990; Cai et al. 2006; Choi et al. 1992; Foissner, Berger, and Schaumburg 1999; Foissner and O’Donoghue 1990; Foissner and Wilbert 1979; Petz, Song, and Wilbert 1995; Sniezek et al. 1991; Snyder and Brownlee 1991; Song and Wilbert 1989; Wasik and Mikozajczyk 1994). Since tintinnid loricae are polymorphic due to environmental factors and the life cycle (Bakker and Phaff 1976; Bernatzky, Foissner, and Schubert 1981; Biernacka 1965; Davis 1978, 1981; Gold and Morales 1975b; Laval-Peuto 1981; Laval-Peuto and Brownlee 1986), the reconstruction of the phylogenetic relationships and hence the establishment of a natural classification require cytological data of a much larger number of tintinnid species. Therefore, Stenosemella pacifica Kofoid and Campbell, 1929 is redescribed in the present paper.
Samples were taken with a 20-μm plankton net from the upper 50 cm of the Irish Sea in front of the Port Erin Marine Laboratory (University of Liverpool) on the Isle of Man in May 2002 at a salinity of about 35‰ and a temperature of about 12 °C.
Cell movement was studied in a Petri dish (about 6 cm across; water depth about 2.5 cm) under a dissecting microscope at about 20 °C. Cell morphology was investigated under a compound microscope equipped with a high-power oil immersion objective as well as bright-field and interference contrast optics. Protargol impregnation followed the protocol of Song and Wilbert (1995). For scanning electron microscopy, cells were fixed for 30 min in a modified Parducz’ solution made of six parts of 2% osmium tetroxide (OsO4, w/v) in artificial sea water and one part of saturated aqueous mercuric-chloride (HgCl2; Valbonesi and Luporini 1990); further steps followed Foissner (1991).
Counts and measurements on protargol-impregnated cells were performed at 1,000X; in vivo measurements were made at 40–1,250X. The kinetal density index is the ratio of kinety number to cell circumference posterior to the membranellar zone [kineties/μm] in protargol-impregnated cells (Snyder and Brownlee 1991). Usually, it was impossible to count all somatic kineties in a specimen as the curved and densely spaced ciliary rows could not be discerned in the laterally orientated fields; hence, the kinetal density index was not calculated.
Attempts to establish cultures at 12–13 °C and a cycle of 12 h dark and 12 h light with an irradiance of about 100 μE/m2/s failed. Furthermore, the sequence analysis of the small subunit ribosomal RNA (SSrRNA) gene was not successful.
Drawing of live specimen summarizes information and is based on mean measurements, while those of protargolimpregnated specimens were made with a drawing device. The kinetal map depicts the ciliary pattern of a protargol-impregnated morphostatic specimen in two dimensions (Choi et al. 1992; Foissner and Wilbert 1979), that is, the cortex is drawn as cut longitudinally along the dorsal kinety; it is also based on mean measurements. Horizontal bars symbolize the collar membranelles, diagonal bars those membranelles that are partially or entirely in the buccal cavity, namely, the elongated collar membranelles and the buccal membranelle. Ratio of cell circumference to length of kineties is 1:1. Kinetids are equidistantly arranged in the ciliary rows, and kinety curvature is neglected, except for the ventral and last kinety whose course might be of taxonomic significance. The somatic cilia are symbolized by perpendicular lines, differences in their length are not considered.
Sampling, re-investigation, and redescription of S. pacifica were conducted by the senior author; the co-author contributed some data.
Slides with protargol-impregnated material, including the neotype, further specimens, and the illustrated dividers, are deposited with the relevant cells marked in the Biology Centre of the Museum of Upper Austria (LI) in A-4040 Linz (Austria). A neotype is provided as (i) no type material is available and (ii) the original description lacks many morphologic and morphometric features. Although the specimens are not from the original type locality (Strait of Georgia, North Pacific), a neotype is established to provide stability in tintinnid taxonomy, following the argumentation of Foissner (2002), Foissner, Agatha, and Berger (2002), and Corliss (2003).
Terminology follows Agatha and Riedel-Lorjé (2006). Two further terms are defined here. Agglutination. The process of accumulation of particles regardless of their source and type, producing an agglomerated lorica. Windows. Holes in the lorica collar or bowl, arranged in one or more horizontal rows. This should replace synonymous terms, such as, “perforations” (Burns 1983), “Löcher” (Brandt 1907), “Maschenlöcher” (Daday 1886), “fenestrae” (Gold and Morales 1976), “Fenster”, “Gitterlöcher”, “Maschen” (Haeckel 1873), “openings” (Wailes 1925; Wang 1936), and “orifices” (Wailes 1943). However, we follow Corliss (1979) as well as Kofoid and Campbell (1929), using the term “window”.
Lorica in vivo 45–60 μm long and 35–45 μm wide, usually composed of an agglomerated bowl and a hyaline collar; 30% of specimens with agglomerated second collar about 40 μm wide and up to 20 μm high (inner diameter not measured, but probably similar to that of hyaline collar; Fig. 1, 2, 9–16, 22–31). Bowl broadly obovate, posterior portion forming an angle of 80°–90°, without process; incrustation very dense, comprising particles of abiotic (silt grains about 10 μm across) and, rarely, biotic (fragments of diatom frustules) origin, matrix and live cell thus invisible. Hyaline collar broadly cylindroidal and with slightly flared rim, encrusted by only few and small particles, with 6–8 windows (estimation as recognizable on only one side of the lorica in scanning electron micrographs); spiralled or annulated structures not recognizable (Fig. 28, 31–33, 36, 37, 39–41). Windows irregularly arranged in slightly constricted proximal collar portion, elliptical to triangular; usually surrounded by a small rim, difficult to recognize in live and protargol-impregnated specimens because only 2.1–3.6 × 1–2 μm( = 2.8 × 1.5 μm; n = 11) in size and often covwith agglomerated particles of the bowl.
Fully extended cells in vivo 55–70 × 15–20 μm, elongate obconical; right side of cell proper gradually merging into the slender, wrinkled, and highly contractile stalk about 7 μm long, attached to bottom of lorica (Fig. 3). Disturbed or preserved cells contracted by about 46% and broadly ellipsoidal with an indentation in the anterior dorsal portion. Macronuclear nodules ellipsoidal, usually in posterior two-thirds of cell proper, each associated with a globular micronucleus; nucleoli 1–2 μm across (Fig. 3, 7). Contractile vacuole and cytopyge not recognizable due to the agglomerated lorica. Myonemes not impregnated. Argyrophilic granules (probably capsules) in striae and tentaculoids. Striae attached to collar membranelles, 0.7–2 × 0.7–1 μm in size, recognizable in live cells and scanning electron micrographs (Fig. (Fig.1,1, ,34).34). Two types of tentaculoids, not recognized in vivo and rarely visible in protargol-impregnations and scanning electron micrographs, possibly because they are contractile: type I originates from outer margin of intermembranellar ridges, finger-shaped, 5–6 × 0.7–1.4 μminsize (Fig. (Fig.33,33, ,36,36, 42, 44); type II originates from inner half of ridges, pin-shaped, up to 9 × 0.5–1.4 μm in size (Fig. 35, 37, 45, 46).
Swimming slowly by rotation about main cell axis, twitching back on obstacles. Disturbed specimens retract quickly into lorica, with motionless membranelles bent to centre of peristomial field; lorica abandonment never observed. When disturbance stops, specimens slowly extending and spreading the collar membranelles almost perpendicularly to main cell axis.
Somatic ciliary pattern of most complex type (Agatha and Strüder-Kypke 2007): comprising a ventral, dorsal, and posterior kinety as well as a right, left, and lateral ciliary field (Fig. 8). Length of kineties and number of kinetids usually highly variable, possibly due to basal body proliferation or resorption in late dividers or postdividers. Ventral kinety commencing 2–3 μm posterior to collar membranelles and anterior to second kinety of right ciliary field, curving slightly leftwards, and extending parallel to kineties of lateral ciliary field, but terminating somewhat posteriorly; composed of monokinetids densely spaced in the anterior portion, but more widely spaced in the posterior, with cilia about 3 μm long after preservation (Fig. 3, 6, 8). Kineties of right ciliary field commencing about 4 μm posterior to collar membranelles, increasing in length from left to right, composed of monokinetids and one anterior dikinetid, except for (i) first kinety starting with 2, rarely 3 dikinetids 2.0–2.5 μm posterior to remaining kineties and terminating near posterior end of ventral kinety and (ii) second kinety occasionally commencing with 2 dikinetids (Fig. 3, 6, 8, 38, 40, ,43).43). Cilia of right ciliary field about 3 μm long after preservation, except for anteriormost cilia of dikinetids (soies; Fauré-Fremiet 1924), which are about 10 μm long in vivo and 7 μm after preservation (Fig. 43). Dikinetidal dorsal kinety commencing 4–6 μm posterior to collar membranelles and about 6 μm apart from left and right ciliary field, extending in distinct leftwards curvature along margin of bulge in posterior portion of preserved cells, and terminating near posterior end of cell proper and posterior kinety, accompanied by argyrophilic granules (Fig. 6–8, ,39).39). Cilia of dorsal kinety associated only with each posterior dikinetidal basal body, in vivo about 10 μm long, and 5–7 μm after preservation. Kineties of left ciliary field commencing about 4 μm posterior to collar membranelles and increasing in length from right to left, composed of monokinetids and one anterior dikinetid (Fig. 6–8, 39, 41); last kineties difficult to distinguish from kineties of lateral ciliary field when peristomial rim hides structure and position of anterior kinetid. Cilia of left ciliary field 1–3 μm long after preservation, except for anteriormost cilia of dikinetids (soies; Fauré-Fremiet 1924), which are about 10 μm long in vivo and 7 μm after preservation. Kineties of lateral ciliary field commencing about 3 μm posterior to collar membranelles (i.e. 0.5–1.0 μm anterior to left and right ciliary field), densely spaced, inclined clockwise, composed of densely spaced monokinetids; first 5–7 kineties about 13 μm long and remaining kineties about 3 μm shorter (Fig. 6, 8, 38, 41). Length of preserved lateral cilia decreasing from about 8 μm at anterior margin of field to 0.5 μm in posterior field portion where, however, also some extraordinarily long (about 20 μm) cilia occur (Fig. 38, 40). Dikinetidal posterior kinety commencing about 4 μm posterior to last or penultimate kinety of left ciliary field, extending almost longitudinally to posterior end of cell proper, and accompanied by argyrophilic granules (Fig. 3, 6, 8). Cilia of posterior kinety associated only with each posterior dikinetidal basal body, about 5 μm long after preservation.
Adoral zone of membranelles closed (Fig. 3, 6–8, 34, 35, 45, 46); occasionally inclined with ventral portion slightly shifted posteriorly. Collar membranelles triangular: cilia increasing in length from about 1 μm at inner end of membranelles to 20–25 μm in outer quarter, and then abruptly decreasing to about 2 μm at the outer end in scanning electron micrographs; distal portion frayed, producing a comb-like appearance (Fig. 34, 35, 39--46).46). Bases (polykinetids) of collar membranelles extend obliquely across peristomial rim, separated by conspicuous ridges (accessory combs) 1–2 μm wide; each about 7 μm long and composed of 3 rows of basal bodies, except for 4 successively elongated bases terminating 3–9 μm posterior to anterior cell end in the buccal cavity (Fig. 3, 8). Single buccal membranelle about 7 μm long and probably composed of 3 rows of basal bodies (Fig. 3). Circular argyrophilic fibre horizontally orientated underneath collar membranelles; its dorsal portion connected with another fibre bundle extending posteriorly (Fig. 18). Proximal ends of elongated collar membranelles and buccal membranelle associated with argyrophilic fibres extending to posterior half of cell (Fig. 3). Endoral membrane extending across peristomial field and right wall of buccal cavity, and composed of a single row of basal bodies, probably with monostichomonad structure; cilia in buccal cavity 2–3 μm long after protargol impregnation, those of distal portion usually covered with perilemma and thus not recognizable in vivo and protargol slides and only rarely in the scanning electron microscope (Fig. 34, 35, 45, 46). Pharyngeal fibres originating from buccal vertex.
Stomatogenesis commences with the apokinetal formation of a roughly cuneate, longitudinally orientated anarchic field of basal bodies in a shallow depression between the end of the ventral kinety and the beginning of the posterior kinety (Fig. 17). While the collar polykinetids commence to differentiate, the oral primordium sinks into a subsurface pouch, enlarges posteriorly, and becomes inverted C-shaped (Fig. 18). The endoral membrane originates de novo as a single row of densely spaced basal bodies. After the arrangement of the membranelles in a closed circle (Fig. 19, 20) and the formation of their cilia, the new oral apparatus evaginates. The division furrow appears just anterior to the opisthe’s adoral zone. The somatic kineties elongate by intrakinetal proliferation of basal bodies and split at the level of the division furrow, usually producing longer ciliary rows in the proter than in the opisthe. A replication band traverses the macronuclear nodules, which then fuse to form an elongate ellipsoidal mass (Fig. 19). After its division, one product each migrates into the proter and opisthe and commences to divide again, reconstructing the initial state of the nuclear apparatus. The parental oral ciliature does not reveal any signs of reorganization during ontogenesis.
Lorica formation was neither observed in live nor in preserved specimens.
Since the cell features of the type population described by Wailes (1925) are unknown, species identification is based on lorica morphology only. According to Laval-Peuto and Brownlee (1986), the diameter of the lorica opening is the least variable dimension and a taxonomically reliable character, whereas lorica length increases during lorica formation and by addition of epiloricae. Actually, the Irish Sea specimens match those from the North Pacific rather well in the diameter of the collar (on average 20 μm vs. 15–17 μm, as estimated from the illustrations provided by Wailes 1925; Fig. 4, 5). The differences in the length and width of the obovate lorica (45–60 × 35–45 μm vs. 35–40 × 30 μm), the angle of the bowl’s posterior portion (about 80°–90° vs. 100°–115°), and the occurrence of an agglomerated second collar in about one-third of the Irish Sea specimens are attributed to the polymorphism of the lorica, especially, as the latter feature was also reported from the congeners Stenosemella ventricosa and S. nucula (Hofker 1931). Thus, we identify our specimens with Stenosemella pacifica.
The loricae of the Irish Sea specimens fall into the size range of the other populations (lorica length: 30–62 μm; bowl width: 24–59 μm; collar diameter: 15–25 μm) and match the angle of the bowl’s posterior portion (70°–115°; Balech 1968; Burns 1983; Kofoid and Campbell 1929; Wailes 1925, 1943; Wang 1936; Xu and Song 2005; Yoo, Kim, and Kim 1988). Furthermore, all populations have windows in the collar; only Xu and Song (2005) did not mention this feature. The number of windows is 8–12, usually 9 (Balech 1968; Burns 1983; Wang 1936; Yoo et al. 1988), while our specimens have only 6–8 windows, as estimated from scanning electron micrographs. The Irish Sea specimens also differ in the occasional occurrence of an agglomerated second collar, a feature not described in the other populations. The variability in the position of the bowl’s greatest width further contributes to its polymorphism. A more detailed comparison with the other populations is impossible, as their cell features are unknown.
The present results and the literature data are insufficient to estimate the entire polymorphism of S. pacifica; culture studies and gene sequence analyses are also required to elucidate the whole intraspecific variability.
There are seven congeners with a similar collar diameter and lorica size: Stenosemella acapulcensis (Osorio-Tafall 1941), Stenosemella avellana (Meunier 1919), Stenosemella lacustris (Foissner and O’Donoghue 1990), Stenosemella monacensis1 (Rampi 1950), Stenosemella nicaraguensis (Osorio-Tafall 1941), Stenosemella nivalis (Kofoid and Campbell 1939), and Stenosemella oliva (Meunier 1910). While S. acapulcensis, S. lacustris, and S. nivalis might belong to different genera due to the lack of windows in the collar, the remaining species cannot clearly be separated from S. pacifica at the present state of knowledge.
There are also some marine Tintinnopsis species with a similar shape and size of the lorica bowl, e.g. T. parva Merkle, 1909 and T. turbo Meunier, 1919; however, they lack a hyaline collar and thus might represent immature loricae of S. pacifica. Dictyocysta ovalis Daday, 1886 apparently differs from our species in the diameter of the short collar (36 μm vs. 15–25 μm), in the nature of the agglomerated particles (coccoliths vs. mineral grains), as well as in the shape (rectangular vs. elliptical to triangular) and size (conspicuous vs. inconspicuous) of the collar windows. Furthermore, the lorica of D. ovalis is longer than that of S. pacifica (72 × 45 μm vs. 30–62 × 24–59 μm; Daday 1886).
Ontogenesis was at least partially studied after protargol impregnation in four species with a similar somatic ciliary pattern: Codonella cratera (Petz and Foissner 1993), Cymatocylis convallaria (Petz et al. 1995), Favella sp. (Laval-Peuto 1994), and Tintinnopsis cylindrica (Agatha and Riedel-Lorjé 2006). Stenosemella pacifica matches these species in the position of the oral primordium posterior to the lateral ciliary field. Furthermore, the somatic kineties of the proter are apparently elongated, while those of the opisthe are shortened compared with morphostatic specimens, indicating a resorption of basal bodies by the proter and a second round of basal body proliferation by the opisthe in late dividers or postdividers (Agatha and Riedel-Lorjé 2006; Brownlee 1983; Petz and Foissner 1993).
Because of the problematic taxonomy, Stenosemella pacifica possibly possesses further synonyms and might have been confused with several species. Especially, the widely distributed species S. nivalis differs from S. pacifica only in the windows (absent vs. inconspicuous and thus easily overlooked). Hence, the distribution of S. pacifica might be much larger.
The records from the temperate North Atlantic (this study), the tropical Gulf of Mexico (Balech 1968), the temperate, subtropical, and tropical North Pacific (Kofoid and Campbell 1929; Wailes 1925, 1943; Wang 1936; Xu and Song 2005; Yoo et al. 1988), and the temperate to subtropical South Pacific (Burns 1983) are substantiated by lorica measurements and illustrations. Additionally, there are uncorroborated records from the Mediterranean Sea (Modigh and Castaldo 2002; Orsi 1936) and the North Pacific (Campbell 1926; Yoo and Kim 1990; Zhang et al. 2002). Stenosemella pacifica co-occurred with similar species (e.g. S. nivalis) at the coasts of Italy (Modigh and Castaldo 2002; Orsi 1936) and New Zealand (Burns 1983). During the sampling period in the Irish Sea (this study), S. pacifica dominated the tintinnid community and was associated with Myrionecta rubra, Tiarina fusus, Laboea strobila, Strombidium capitatum, Cyrtostrombidium sp., Strombidinopsis acuminata, Pelagostrobilidium neptuni, and several Rimostrombidium species.
Wailes (1925) established Tintinnopsis punctata forma minor. Kofoid and Campbell (1929) regarded the form as distinct species within the genus Stenosemella. They introduced the replacement name S. pacifica, arguing that the specific epithet “minor” was occupied by Tintinnopsis ventricosa var. minor Fauré-Fremiet, 1908 and T. ventricosa var. minor Rossolimo, 1927, which are both synonyms of S. nivalis.
By raising the forma minor established by Wailes (1925) to species rank in the genus Stenosemella, Kofoid and Campbell (1929) would not have created a secondary homonymy. Hence, the introduction of a replacement name was unnecessary. The correct name according to the Principle of Priority (article 23.1.; ICZN 1999) would be Stenosemella minor (Wailes, 1925) Kofoid and Campbell, 1929. However, this name was never used, even not by Wailes (1943), while the replacement name introduced by Kofoid and Campbell (1929) was cited in at least 18 articles during the past 68 yr. Although the requirements of the reversal of precedence (article 23.9.; ICZN 1999) are thus not fulfilled concerning the number of citations and the time span, we pledge to consider S. pacifica Kofoid and Campbell, 1929 a nomen protectum for the reasons of nomenclatural stability (article 23.2.; ICZN 1999).
Lorica on average 40–50 μm long and 30–40 μm wide, broadly obovate; occasionally with agglomerated second collar. Hyaline collar about 20 μm wide, with 6–8 windows. Extended cell on average in vivo 45–60 × 15 μm, elongate obconical, and highly contractile. Two macronuclear nodules and two micronuclei. Ventral kinety commences anterior to second kinety of right ciliary field. About 5 kineties each in right and left ciliary fields, all composed of monokinetids and 1 anterior dikinetid, except for second kinety with 2 or 3 anterior dikinetids and third kinety with occasionally 2 dikinetids. Lateral ciliary field with about 11 monokinetidal kineties. On average 30 dikinetids in dorsal kinety and 6 in posterior kinety, with a cilium only at each posterior basal body. About 18 collar membranelles of which 4 extend into buccal cavity; 1 buccal membranelle.
Pelagial of the Irish Sea near the village of Port Erin, Isle of Man, United Kingdom (54°05′06″N, 04°45′50″W).
Genus Stenosemella Jörgensen, 1924
Lorica firm, composed of an agglomerated bowl and a small hyaline collar with a single row of inconspicuous windows at collar base. Somatic ciliature comprises a ventral, dorsal, and posterior kinety as well as a lateral, right, and left ciliary field.
Tintinnus ventricosus Claparède and Lachmann, 1858 (subsequent designation by Kofoid and Campbell 1929).
Kofoid and Campbell (1939) established the genus Luminella with Tintinnopsis punctata Wailes, 1925 as type species. The genus was distinguished from Stenosemella by the presence of small windows in the collar. Balech (1959), as well as Gold and Morales (1976), however, revealed also windows in the collar of S. ventricosa (Claparède and Lachmann, 1858) Jörgensen, 1924, the type of the genus. Hence, the genus Luminella is a synonym of Stenosemella, as already suggested by Yoo et al. (1988), and the affiliated species (Luminella inflata, Luminella pacifica, and Luminella punctata) are retransferred to the genus Stenosemella. Likewise, Luminella neocalifornica is affiliated with the genus Stenosemella as S. neocalifornica (Osorio-Tafall, 1941) nov. comb.
Besides S. ventricosa and S. pacifica (this study), six congeners possess windows in the collar: Stenosemella epunctata Wang, 1936; Stenosemella inflata Kofoid and Campbell, 1929; Stenosemella neocalifornica (Osorio-Tafall, 1941) nov. comb.; Stenosemella punctata Wailes, 1925; S. oliva (Meunier, 1910) Kofoid and Campbell, 1929 according to the scanning electron microscopic study by Gold and Morales (1976); and Stenosemella steini (Jörgensen, 1912) Jörgensen, 1924 according to the scanning electron microscopic study by Gold and Morales (1975a). However, there are also three Stenosemella species with fully developed collars that definitely do not have windows: S. acapulcensis Osorio-Tafall, 1941; S. lacustris Foissner and O’Donoghue, 1990; and S. nivalis (Meunier, 1910) Kofoid and Campbell, 1929 according to the revision by Kofoid and Campbell (1939). Because the life cycle of Stenosemella species is unknown, it cannot be excluded that they might form loricae with and without windows; Burns (1983), for instance, speculated that S. pacifica and S. nivalis might represent ecotypes or growth/age stages of the same species due to the overlapping distribution. Hence, we refrain from establishing a distinct genus for S. acapulcensis and S. nivalis at the present state of knowledge. Likewise, the freshwater species S. lacustris with its agglomerated collar should be excluded from the genus; however, a proper affiliation with the genus Tintinnopsis is impossible as the cell features of T. beroidea Stein, 1867, the type of the genus, are unknown. Additionally, there are nine Stenosemella species in which the inconspicuous windows were not mentioned: S. avellana (Meunier, 1919) Kofoid and Campbell, 1929; S. bernhardi Busch, 1948; S. brevicolli Busch, 1948; S. compressa Busch, 1948; S. expansa (Wailes, 1925) Kofoid and Campbell, 1929; S. monacensis Rampi, 1950; S. nicaraguensis Osorio-Tafall, 1941; S. perpusilla Hada, 1970; and S. producta (Meunier, 1919) Kofoid and Campbell, 1929. Tintinnopsis rioplatensis Souto, 1973 is apparently closely related to these species as it occasionally forms a hyaline collar, which is 5 μm high and possibly without windows (Souto 1973).
Although the phylogeny of tintinnids is widely unknown, recent cladistic and gene sequence analyses indicated a paraphyletic origin of the hyaline loricae (Agatha and Strüder-Kypke 2007; Strüder-Kypke and Lynn 2003). Accordingly, the lorica structure seems to be less important for the higher tintinnid classification, while it remains a reliable feature at generic level. Therefore, we do not change the familial affiliation of Stenosemella at the present state of knowledge, although it might be transferred in the future from the family Codonellopsidae Kofoid and Campbell, 1929 to the family Dictyocystidae Haeckel, 1873.
Besides Stenosemella, there are three genera having loricae composed of an agglomerated bowl and a hyaline collar that at least occasionally comprises windows: Codonellopsis Jörgensen, 1924; Dictyocysta Ehrenberg, 1854; and Wangiella Nie, 1934. Stenosemella differs from the former genus in length and structure of the collar (short vs. long; usually without vs. with spirals); however, it cannot be excluded that a spiralled collar (epilorica; Laval-Peuto and Brownlee 1986) might be added to the short collar of Stenosemella species during the life cycle. Stenosemella is distinguished from Dictyocysta and Wangiella by the size of the collar windows relative to the collar size (windows inconspicuous, restricted to the collar base, and with mullions usually broad relative to windows vs. conspicuous, occupying almost entire collar, and with mullions usually slender relative to windows). To our knowledge, transitions between these two types of windows do not exist.
The somatic ciliary pattern is included into the genus diagnosis, although the pattern of the type species was only provided in Pierce’s (1996) unpublished PhD Thesis. However, the same pattern was observed in S. pacifica (this study), S. oliva (Pierce 1996), and S. steini as inferred from illustrations provided by Laval-Peuto (1994), Lynn and Small (2002), as well as Small and Lynn (1985). Members of the genera Codonella Haeckel, 1873; Codonellopsis Jörgensen, 1924; Cymatocylis Laackmann, 1910; and Tintinnopsis Stein, 1867 share this complex ciliary pattern (for review see Agatha and Strüder-Kypke 2007). According to illustrations of protargol-impregnated specimens, Climacocylis Jörgensen, 1924 and Protorhabdonella Jörgensen, 1924 apparently also possess a similar arrangement of the somatic ciliature (Lynn and Small 2002; Small and Lynn 1985). Further investigations are required to decide whether the subtle differences recognizable in the structure, position, and curvature of the kineties are genus- or only species-specific.
Stenosemella crateri is frequently mentioned in the internet. According to the authors’ knowledge, such a species was not described; hence, it is considered a nomen nudum.
The taxonomic investigations by S. Agatha were supported by the Austrian Science Foundation (FWF; Projects T 116 and P17752-B06). The stay of Sheng-Fang Tsai in Salzburg, Austria, was supported by the National Science Council, Taiwan, Republic of China (Grant 096-2917-1-019-001). Many thanks are due to Wilhelm Foissner (Organismal Biology, University of Salzburg, Austria) for his constructive criticism. Thanks are also due to David Montagnes (School of Biological Sciences, University of Liverpool, UK) for placing his laboratory on the Isle of Man at the senior author’s disposal.
1STENOSEMELLA is feminine gender (Aescht 2001); thus, the specific epithet is changed: Stenosemella monacensis Rampi, 1950 nom. corr.