Contributions to Zoology, 68 (3) -.. (1999)
Scanning electron microscopy of acrothoracican cypris larvae (Crustacea, Thecostraca, Cirripedia, Acrothoracica, Lithoglyptidae)
Gregory A. Kolbasov , Jens T. Hoeg , Alexei S. Elfimov
Keywords: Cirripedia, Acrothoracica, Ascothoracida, cypris larva, morphology, lattice organ, SEM, phylogenetic relationships, larval characters
Scanning electron microscopy was used to provide a full morphological description of cypris morphology in the acrothoracican species Lithoglyptes mitis and L. habei (Lithoglyptidae). Special attention was given to lattice organs, antennules, thorax, thoracopods, abdomen, and furcal rami. Cypris larvae of the Acrothoracica share some putative plesiomorphic features with the cypris-like ascothoracid larvae of the non-cirripede taxon Ascothoracida. The most notable are traces of abdominal segmentation and carapace lattice organs without pore fields. Acrothoracican cyprids also share numerous synapomorphies with those of the Thoracica and the Rhizocephala. This list includes a four-segmented antennule with a triangular first segment of two sclerites set at an angle to each other, a cylindrical second segment, a small third segment functioning as an attachment organ, and a cylindrical fourth segment bearing homologous sensory setae. Further apomorphies are a pair of fronto-lateral horn glands exiting antero-ventrally on the head shield (carapace), a pair of multicellular cement glands exiting on the attachment organs, a single stout, serrated and non-natatory seta on the thoracopodal exopods and a highly reduced abdomen with at best traces of segmentation. These synapomorphies in cypris morphology support a monophyletic Cirripedia comprising the Acrothoracica, Thoracica, and Rhizocephala but excluding the Ascothoracida.
The Cirripedia consist of the orders Acrothoracica, Thoracica, and Rhizocephala (Høeg, 1992; Høeg, 1995). Both the Thoracica and the Acrothoracica use thoracic limbs (cirri) for setose feeding, but the Acrothoracica deviate in inhabiting burrows and lacking an armament of mineralized shell plates. This cryptic mode of life has resulted in numerous modifications to their morphology.
The Acrothoracica have long been important in discussing both the evolution and phylogeny of the Cirripedia and of the Thecostraca in general (Glenner et al., 1995; Newman 1971, 1974, 1987; Spears et al., 1994; Turquier, 1972). An origin from or, more precisely, a sister group relationship with various cirripede taxa has been suggested, such as the Iblidae (Tomlinson, 1969) and the scalpellid Lithotrya (Newman, 1982; Grygier & Newman, 1985). In spite of these studies the position of the Acrothoracica within the Thecostraca remains unclear. The lack of mineralized shell plates and whether this is primary or due to secondary loss impedes comparison with the characters used in estimating the phylogeny of thoracican barnacles (Glenner et al., 1995, Newman 1996). Spears et al. (1994) used molecular data in an effort to clarify cirripede phylogeny and their results suggested an affinity between the Acrothoracica and the Ascothoracida, which challenged even the basic monophyly of the Cirripedia. Despite the wide morphological differences among adult Thecostraca, most representatives of each group possess pelagic nauplius and cypris-like larvae. The similarity and unproblematic homology of these larvae including numerous details apparent under the scanning electron microscope make them eminently suited for resolving phylogenetic the relationships within the Thecostraca (Grygier 1987a,b; Walossek et al., 1996). Jensen et al. (1994) used SEM on cypris larvae to study the recently discovered lattice organs in the carapace. They found putative plesiomorphic similarities between Ascothoracida and Acrothoracica and putative synapomorphies between Acrothoracica and the remaining two cirripede orders. Moyse et al. (1995) employed SEM on cypris attachment organs in a wide selection of cirriped cyprids. These two studies focussed on but a few selected organs, and all previous works on acrothoracican cyprids used only light microscopy (Kühnert, 1934; Tomlinson, 1969; Turquier, 1967, 1970, 1971, 1985; Wells & Tomlinson, 1966). For acrothoracican cyprids we therefore lack the level of detail now available from the two other cirripede orders (e.g., Elfimov 1995; Glenner et al., 1989; Glenner & Høeg, 1995; Høeg, 1985; Jensen et al., 1994; Moyse et al., 1995; Walker, 1985; Walker et al., 1987). An ultrastructural study of the entire morphology of an acrothoracican cyprid is necessary for gathering the suite of larval characters that Glenner et al. (1995) advocated for future studies of cirripede phylogeny. Here we try to accomplish this in part by a study of cypris morphology in two species of the acrothoracican family Lithoglyptidae.
Material and methods
Most acrothoracican species brood their larva until the cypris stage is reached (Tomlinson, 1969). It is therefore possible to sample mature cypris larvae from the mantle cavities of females found in museum collections. Acrothoracican males are always dwarf forms attached to the females (Gotelli & Spivey, 1992; Kolbasov, 1996).
We examined the collections of mollusc shells stored in the Zoological Museum of Moscow State University and found more than 300 specimens containing Acrothoracica. Many of them hosted either Lithoglyptes habei or L. mitis, and some contained cypris larvae. Recently settled cypris stages of dwarf males (Fig. 2C) and females were also isolated.
All material was preserved in 70% alcohol. We investigated five Lithoglyptes habei cypris larvae and five L. mitis cypris larvae with SEM and we also mounted some for light microscopy after KOH treatment. All larvae for SEM investigation were post-fixed with 2% OsO4 for 2hrs, dehydrated in acetone, and critical point dried in CO2. Dried specimens were sputter-coated with gold and examined at 15 kv accelerating voltage (with a JEOL JSM-840 SEM in Copenhagen and a HITACHI S405A SEM in Moscow). After investigation of external carapace features one “valve” of some larvae was removed to reveal the body.
The material studied came from the following localities. Lithoglyptes habei: Gulf of Aden, 13o59‘5‘‘N, 48o24‘7‘‘E, depth 3m, coral reef, 1 female with 1 cypris inside, in Turbo argirostomum; Seychelles, Silhouette I., 4o36‘S, 56o48‘E, subtidal zone, 6 females and 1 free cypris in Mancinella mancinella; South China Sea, Vietnam 12oN, 109oE: depth 1.5 m, 3 females (1 with a cypris inside) in Mancinella mancinella; depth 2 m, 2 females and 1 free cypris in Coralliophila deformis; 2-4 m, 5 females (1 with a cypris inside) in Drupa morum.
Lithoglyptes mitis: Maldives: Feartu I., 3o48‘N, 73o05‘E, intertidal zone, coral reef, 8 females (1 with a cypris inside); Genego I., 3o49‘N, 73o06‘E intertidal and subtidal zones, coral reef, 2 females and 1 cypris with stretched antennules in Trochus pyramis, 17 females (2 with a cypris inside) in Mancinella alauina, 8 females (1 with a cypris inside) in Latirolagena smaragdula, 3 female specimens (1 with a cypris inside) in Morula cavernosa, 2 females (1 with a cypris inside) in Hipponix sp.
We could not detect any morphological differences between cypris larvae of Lithoglyptes habei and L. mitis. We have therefore not distinguished between the two species in the following description, but the species name is provided for all figures.
Size and shape of the cypris
In our material the adult females never contained more than a single cypris, while Tomlinson (1969) often found two brooded larvae in the same female. The brooded cypris larvae are located in the lower body part of the female and usually towards the dorsal margin of mantle cavity. They measure ca. 580-600 µm in length, which equals approximately 40% of the length of the brooding female. The size of the cypris relative to the female means that brood size is necessarily minute.
The cypris head shield, or carapace, has an elongated, spindle-shaped form with an anterior rounded end and a narrower and truncated posterior end (Figs. 1B,C, 2A). The dorsal margin is slightly curved. The ventral margins of the carapace are also slightly curved in the anterior half, whereas in the posterior half they are somewhat concave. The length:height ratio of the carapace is ca. 3:1. In lateral view the shape of the cypris resembles those of other cirripedes. However, the cyprids of both Lithoglyptes studied here and those of Trypetesa lampas studied by Jensen et al. (1994) deviate from other cirripede cyprids in having a very narrow carapace compared to the length (Fig. 2A,B). The functional significance of this shape is not clear. In Trypetesa it may relate to peculiarities of settlement within the narrow confines of gastropod shells inhabited by hermit crabs.
Fig. 1. A - Lithoglyptes mitis, adult female, lateral view, with a cypris (bottom arrow) and an unidentified copepod (top arrow) inside the mantle cavity, lateral view; B,C - L. habei cyprid, lateral view from light microscopy; D - terminal part of antennule showing attachment disc on 3rd segment (3) with dense carpet of cuticular villi (see Fig. 5D); E - Setation on fourth antennulary segment, among five terminal setae (TS) only minute seta C identified (See Fig. 6D); F - Setation (arrowheads) of terminal (2nd) segment in thoracopodal exopod, isolated basal seta, three subterminal and two terminal setae closely grouped; G - Setation (arrowheads) of terminal (3rd) segment of thoracopodal endopod, three terminal setae; H - Terminal end of furcal ramus with two setae (arrowheads) (See Fig. 7G). CE compound eye, CG cement gland, CT, rows of ctenes inside mantle cavity, FR furcal ramus, LO1-5 lattice organs 1-5, M extrinsic antennulary muscles, NE nauplius eye, OC, oral cone, PAS postaxial sensillum, PS2 postaxial seta 2, RS radial setae, SC proximal sclerite of 1st antennulary segment, SK skirt encircling attachment disc, SS1-4 subterminal setae 1-4, TE telson, TH thorax, THP thoracopods, TS terminal setae, 1-4 antennulary segments. Scale bars in µm.
The general morphology of Lithoglyptes cyprids agrees with that found in other cirripedes. The four-segmented antennules are located in the anterior mantle cavity, which occupies the anterior one third of cyprid carapace. The cement glands lie at the basis of the first antennulary segments. They have a loboform shape and a brown color (in alcohol), darker than the rest of the body. The paired compound eyes, associated with frontal filaments, lie in front of the cement glands (Fig. 1B,C). The nauplius eye is situated near the dorsal margin, one third of the length from the anterior end. Retractor muscles of the antennules and thorax extend through the anterior and middle parts of the cypris body and attach to the dorso-medial side of the carapace, but a complete description of the cypris musculature will require section series as in Walley (1969) and Høeg (1985). The undifferentiated oral (buccal) cone and the thorax lie within the posterior half of the carapace. The thorax carries six pairs of biramous limbs armed with long natatory setae. The abdomen is rudimentary but carries a distinct telson with a pair of furcal rami (Fig. 1B, 7). The thorax and the antennules can be partially extended outside the mantle cavity (Figs. 2B,D)
There is no dorsal hinge line or posterior slit on the carapace. At lower magnifications its surface appears slightly wrinkled with small longitudinal and transversal ridges (Fig. 2A,B), but some of these ornaments may be artifacts produced during fixation or preparation for SEM. At higher magnification the carapace surface appears smooth (Fig. 3A-D) and without the cellular hexagonal patterning that characterizes some ascothoracid larvae (Itô & Grygier, 1990), facetotectan cypris-y (Schram, 1970), and some thoracican cyprids. The carapace also lacks large pores (except the fronto-lateral pores), papillae, and the wheel organs of Elfimov (1995). Unlike cyprids of Cryptophialus (unpublished data), there are no long seta but only minute (0.7 µm) setae sparsely distributed over the entire carapace (Fig. 3A,C,D). These carapace setae are single (Fig. 3A) or double (Fig. 3C,D) and are located in shallow, 1 µm wide depressions. Single setae occur most frequently on the anterior and lateral surfaces of the carapace. A longitudinal row of double setae extends from the anterior end along the dorso-medial line of the carapace (Fig. 3C,D). The carapace “valves” have deep longitudinal and transverse furrows at the anterior end (Figs. 2D, 3C).
Lithoglyptes cypris larvae lack the fronto-lateral horns found in a few thoracican species, but sport a pair of conspicuous fronto-lateral pores near the antero-ventral margin of the carapace ca. 80 µm from the anterior end (Figs. 2D, 3B,E). The pores are elongate (3 by 4 µm) and surrounded by a 1.3 µm high cuticular ridge without any sculpture. They are the exits for the fronto-lateral glands that characterize cirripede cyprids.
The carapace has five pairs of lattice organs (LO) in both Lithoglyptes habei and L. spinosa. They have the same position and the same morphology of the individual organs as the five LO pairs found in Trypetesa lampas by Jensen et al. (1994).
Individual morphology. All individual lattice organs are shallow, 7-18 µm long and 0.8-1 µm wide depressions (Fig. 4). They may have a weak median keel or crest (in LO3 and LO4), but are never encircled by a cuticular ridge. This means that the LOs of Lithoglyptes have the same morphology (‘keel in a trough‘) as found by Jensen et al. (1994) in cyprids of T. lampas. The elongate depression of the LOs has a latticed bottom due to minute perforations in the epicuticle, but a TEM investigation of similar organs in the acrothoracican T. lampas revealed that they lack pores in the underlying procuticle (Høeg et al., 1998).
Terminal pore. A conspicuous terminal pore (TP) lies at the posterior end of LO1 and LO3-5, but LO2 differs in having this pore sited at the anterior end (Fig. 4A-D).
Position and shape. The first (LO1) and second (LO2) pairs of lattice organs lie close (10 µm) together, ca. 200 µm from the anterior end and 5-13 µm from the dorsal midline (Fig. 4A,B). LO1 is straight and 7 µm long. LO2 is longer (18 µm) and very slightly curved. The three posterior pairs (LO3-5) are located ca. 300 µm behind the anterior pairs and are close (100 µm) to the posteriormost end of the cypris (Fig. 4E). LO3 and LO4 are separated by only 4 µm and lie approximately in line with LO1 and LO2 (4-5 µm from the dorsal midline). LO5 lie level with LO4 but are more distant (16-18 µm) from the midline (Fig. 4E). LO3 and LO4 are 8-10 µm long and straight or only slightly curved. LO5 is only 7 µm long.
Mantle and mantle cavity
The inner surface of the mantle has a longitudinal fold 18 µm distant from the ventralmost edge (Fig. 5C, white arrowheads). In the anterior mantle cavity, the cuticle also has five longitudinal rows of cuticular ctenes running parallel to the ventral edge of the carapace. These rows are separated by about 11 µm apart and their 3-4 µm high ctenes consist of unfused fringes (Figs. 1C, 5A,B). We were not able to verify whether similar ctene rows occur in the posterior part of the mantle cavity (Terminology on ctenes and fringes adapted from Grygier (1988) and Klepal & Nemeschkal (1995)).
1st antennulary segment
As in all other cirripede cyprids the antennules consist of four segments (Figs. 1B, 2D, 5C,D, 6). The first segment (Figs. 1C, 5C) is largest, about 80 µm long and 40 µm wide, flattened laterally and without setae. As in all cirripedes this segment has a conical or triangular shape and consists of two sclerites (see detailed description in Høeg, 1985 and Glenner, in press). The proximal sclerite forms the base of the cone and carries two anteriorly projecting rods (SC in Fig. 5C). The distal sclerite connects with cylindrical second segment at the apex of the cone. The segment can be completely withdrawn inside the mantle cavity, but projects outside during exploratory walking and in attachment (Figs. 1C, 2C).
2nd antennulary segment
The cylindrical second segment is ca. 75 µm long, with a straight ventral (postaxial) margin and curved dorsal (preaxial) margin (Fig. 5C). Its width decreases from 35-40 µm proximally to only 12-13 µm at the articulation to the third segment. A conspicuous postaxial seta 2, number 13 in the classification of Nott and Foster (1969), inserts ventrodistally (Figs. 1D, 5D) as in the groundpattern for cirripede cyprids (Glenner et al., 1989; Moyse et al., 1995). There are no other second segmental setae, but a small ctene (7-10 fringes) adorns its lateral surface closer to the basal margin than to the distal one.
3rd antennulary segment
The hoof-shaped third segment measures ca. 30 by 15 µm (Figs. 1D, 5D, 6A,C). All surfaces except the attachment disc lack setae. A well developed skirt of thin cuticle (sensu Moyse et al., 1995) encircles the attachment disc, which is morphologically ventral (Figs. 5D, 6A,C) and is covered by a dense carpet of cuticular villi (Figs. 1D, 6A). We failed to detect an axial sense organ (sensillum), but in other cyprids it is often obscured by the cuticular villi (Moyse et al. 1995). A well-developed postaxial sensillum (PAS) inserts at the postaxial margin of the segment (Figs. 1D, 6A) between the skirt and the carpet of villi. Two setae situated at the distal, preaxial margin of the attachment disc (Figs. 1D, 5D) represent two of the radial sensilla of thoracican cyprids, but there could well be additional radial sensilla hidden in the carpet of villi. An indistinct structure inserts on the lateral surface near the base of the fourth segment (Fig. 6B). We interpret it as a rudimentary postaxial seta 3, which in thoracican and some rhizocephalan cyprids inserts at this place but is a very conspicuous sensillum (Nott & Foster, 1969; Moyse et al., 1995).
4th antennulary segment
The cylindrical fourth segment measures 5 µm by 4 µm and inserts laterally on the third segment (Figs. 5F, 6A-D). As in thoracican cyprids (Clare & Nott, 1994), the segment carries four subterminal and five terminal setae (Figs. 1E, 6D). Our terms of these setae follows the scheme of Gibson & Nott (1971) and also used by Clare & Nott (1994), Glenner & Høeg (1995) and Walossek et al. (1996). All setae of the fourth segment, except the diminutive seta C, have an apical pore.
The four subterminal setae (setae 1-4) are situated close together. They are morphologically identical, approximately equal in length (9 µm), and resemble those in thoracican barnacles (SS1-4 in Fig. 4D).
The homologies of the five terminal setae are discussed below. They differ in regards to length, width, and morphology, but not to the extent seen in the Thoracica (Gibson & Nott, 1971; Clare & Nott, 1994; Glenner & Høeg, 1995). In particular, all terminal setae in Lithoglyptes are unsetulated, whereas in thoracicans two terminal setae (A & B) have long setules. In Lithoglyptes, two setae are 8-9 µm in long, narrow, and of simple form. One seta (C) is vestigial, just as in the Thoracica. The longest seta (13 µm) has a distended basal half, but tapers rather abruptly at around 2/3rds of the way towards the tip; its surface shows a faint circular or spiral pattern that indicates a reinforcement structure in an otherwise very thin cuticle. Finally, one c. 9 µm long, cylindrical and isodiametrical seta has a width in between the two thin setae and the thick seta and terminates very abruptly.
Thorax and thoracopods
The ca. 120 µm long thorax forms the posterior third of the cypris body (Fig, 1A,B, 7). As in other cirripedes it consists of six segments each bearing a pair of biramous and natatory thoracopods. In rhizocephalan cypris larvae Walossek et al. (1996) found an unpaired medio-ventral process inserted between the sixth thoracic segment and the abdominal rudiment, and they speculated that it might be a rudimentary penis. The acrothoracican cypris has no such process (Fig. 7D,H).
We could not observe all details in each of the six individual pairs of thoracopods. Small differences, such as occur between the first and second pair in the Thoracica may therefore have gone undetected (Glenner & Høeg, 1995). The thoracopods in cyprids of Lithoglyptes resemble those described using light microscopy from cyprids of Trypetesa nassaroides (Turquier, 1967). The thoracopods consist of a protopod carrying a two segmented exopod and a three segmented endopod (Fig. 7C,D). In lateral view, sub-quadrangular sclerites (Fig. 7D) cover the insertion and the proximal part of the thoracopods, so we could not with certainty identify a coxa. The thoracopods bend strongly backwards in the joint between the first ramal segment and the basis before they insert on the thorax (Fig. 7D,F).
The first exopod segment carries a single, stout seta inserted latero-distally beneath a triangular projection and extending beyond the second segment (SES in Fig. 7D). It carries a basal row of small spines and a single row of 8-10 much larger spines on the surface facing the second segment. Extensive bending can occur between the first and second exopodal segments. The second segment bears five terminal simple setae and a single isolated simple seta inserted near the base (Figs. 1F, 7C,D).
The endopod is shorter than the exopod. The position of segments in preserved cyprids show that extensive bending can occur between segments two and three. We never saw any flexure
between segments one and two in the fixed specimens and surmise that in the live larva little or no movement occurs between them at all (Fig. 7D). The setation of the endopod is difficult to observe due to obscuration by the exopods, but the second segment seems to carry one simple seta on the latero-distal margin, while the third segment (Figs. 1G, 7C) carries three simple, terminal setae. If correct, this setation corresponds exactly to that described by Turquier (1967) for the endopod of Trypetesa nassaroides cyprids.
Hindbody and furcal rami
The hindbody consists of a short, cylindrical (6 x 10 µm) abdomen, with four transverse furrows on the ventral side, that may indicate the presence of four segments, and a longer (18 x 13 µm) telson (Fig. 7A,D,F).
The lateral surface of the telson bears a row (ctene) of five elongate denticles (Fig. 7H). In addition there is a single, minute seta dorso-distally on each side of the telson and a small group of setae or fringes laterally on the ventro-distal margin (Fig. 7H). The ventral surface of the telson has a deep and distinct medial cleft (Fig. 7F).
The furcal rami in cyprids of Lithoglyptes have only a single 44 µm long segment (Fig. 7A,E,G). However, the partially cleaved telson can easily be mistaken as the basal segments of “two-segmented rami”. The lateral surface of a caudal ramus bears a ctene of nine fringes and some sparsely spaced smaller fringes (Fig. 7E), while distally it terminates in two long setae (Fig. 1H). A comb of cuticular villi surrounds the furcal setae on the dorsal margin, but are almost vestigial along the ventral margin (Fig. 1H, 7G).
This paper is the first SEM based description of all external features in acrothoracican cypris larvae. Jensen et al. (1994) and Moyse et al. (1995) have previously used SEM to describe individual organs in cyprids of Weltneria spinosa (lattice organs) and Trypetesa lampas (lattice organs, attachment organs). In addition, Turquier (1967) gave a good light microscopical account of cyprids of T. nassaroides.
The cyprids of Lithoglyptes agree in all important aspects with the scattered data available for the other acrothoracican species. We can confirm that acrothoracican cyprids show numerous similarities with cyprids of the remaining two cirripede orders while also in many respects resembling the cypris-like ascothoracid larvae of the Ascothoracida. We will first discuss the individual morphological features concerned before we summarize their phylogenetic significance.
Lattice organs were first described by Elfimov (1986). They are chemoreceptors and known only from within the Thecostraca, where five pairs adorn the carapace of facetotectan cypris-y larvae and cirripede cypris larvae (Jensen, 1993; Jensen et al., 1994; Høeg et al., 1998; Hosfeld et al., 1998). Many ascothoracidans have fewer lattice organs (Grygier, pers. comm.), but based on comparison with the fixed number in both the Facetotecta and the Cirripedia we consider such cases as apomorphic deviations from a thecostracan ground pattern with five pairs. Lattice organs (LO) vary between the thecostracan taxa both in their general morphology and in the position of the large terminal pore, which can be sited either anteriorly or posteriorly in the individual organ (Jensen et al., 1994).
Lithoglyptes habei, L. mitis, Trypetesa lampas, and Weltneria spinosa have lattice organs of near identical morphology (“keel in a trough” type) and in all of them the terminal pore is anterior in LO2 but posterior in LO1 and LO3-5. This stereotyped morphology of the lattice organs within the Acrothoracica makes them well suited for discussing large scale thecostracan phylogeny. The Acrothoracica have lattice organs of the “keel in a trough” morphology, but all the numerous thoracican and rhizocephalan species studied by Jensen et al. (1994) had lattice organs of the “pore field type”, i.e., an oval or elongate area perforated by numerous small pores. This variation in lattice organ morphology found among the Cirripedia can be polarized by outgroup comparison with the Ascothoracida and the Facetotecta. Both outgroups have lattice organs of the keel “in a trough” type (Jensen et al., 1994; Hosfeld et al., 1998), which indicates that this morphology is plesiomorphic within the Thecostraca. The apomorphic “pore field” type shared by cypris larvae of the Rhizocephala and the Thoracica indicates that these two taxa are sister groups, a conclusion also supported by molecular data (Spears et al., 1994).
The variation in the position of the large terminal pore is more problematic. It is posterior in all five pairs (LO1-5) in the ascothoracidan Ulophysema oeresundense, whereas the Thoracica and Rhizocephala have an anteriorly sited pore in LO1 and LO2 but a posteriorly sited pore in LO3-5. Jensen et al. (1994) therefore suggested that the Acrothoracica exhibits an intermediate character state in having an anteriorly sited pore in LO2 only. There is nothing in our data to question that conclusion. But Grygier & Ohtsuka (1995) found an anterior position of the terminal pore in LO1 and LO3 of Synagoga millipalus indicating the different ascothoracidan species vary with respect to the position of the terminal pore. We therefore need a third group to clarify the plesiomorphic position of the terminal pores within the Thecostraca. This highlights the importance of a detailed SEM investigations of the cypris-y larvae of the Facetotecta (Hosfeld et al., 1998).
Parallel longitudinal rows of cuticular ctenes line the inner lamellae of the carapace in both acrothoracican cyprids (present study), rhizocephalan cyprids (Walossek et al., 1996), thoracican cyprids (Høeg, unpublished) and ascothoracid larvae (Grygier, 1988; Ito & Grygier, 1990). This indicates that such rows represent a ground pattern (apomorphic?) feature of the Thecostraca. We suggest that they function as combs for cleaning the antennules.
Segments 1 and 2 has a morphology very similar to that seen other cirripede cyprids. Segment 3 resembles the one Moyse et al. (1995) described from Trypetesa lampas, but with the differences noted below. In all cirripede cyprids, the radial setae are partially or wholly obscured by the microvilli carpeting the attachment disc, so an exact count requires TEM sections in a plane tangential to the disc. This has only been done in Semibalanus balanoides, where Nott & Foster (1969) found 8 radial setae. It is quite normal that two of the radial setae distally on the attachment disc are longer than the remaining ones (Moyse et al., 1995), as also seen here in Lithoglyptes, where they are the only ones visible at all. As in our study, Moyse et al. (1995) failed to detect the axial sense organ in Trypetesa lampas using SEM only but revealed its presence with TEM. An axial sense organ could therefore also be present in our species.
In T. lampas Moyse et al. (1995) specifically mentioned the absence of a postaxial seta 3 (ps3), which in both the Thoracica and Rhizocephala is a conspicuous structure that inserts near the base of the 4th segment. In Lithoglyptes cyprids the rudimentary knob found adjacent to the insertion of segment 4 could represent a highly a reduced ps3. If so, this indicates that the absence or extreme reduction of ps3 is an apomorphy for the Acrothoracica.
Antennulary segment 4
This segment has the same number and position of setae as in thoracican cyprids. The homology of the four subterminal setae with the similarly shaped and positioned ones in thoracican cyprids is straightforward. There is also little doubt that the five terminal setae correspond to the five terminal ones in the Thoracica. One terminal seta (C) is very short in both the Thoracica and Acrothoracica and clearly homologous in both groups. The remaining four much longer terminal setae differ in morphology between the two orders so we hesitate to suggest any seta-by-seta homologies. The seta in Lithoglyptes with a distended basal part and tapered apex could well correspond to seta D in thoracicans (and rhizocephalans). In both acrothoracicans and thoracicans this seta exhibits a distinct pattern on its surface and Clare & Nott (1994) suggested that it is an aesthetasc.
Apomorphies in antennulary morphology
The unique structure and morphology of the cypris antennule involve several putative apomorphies which agrees with the claim that the Acrothoracica, Thoracica and Rhizocephala form a monophyletic Cirripedia. In all three orders, the antennule consist of four segments with surprisingly similar shape and function. This probably reflects functional constraints posed on the antennule that functions both in exploratory walking prior to settlement and in permanent cementation. Important apomorphies are: a triangular or cone-shaped first segment consisting of two sclerites set at an angle to each other (see Høeg 1985; Glenner in press); a long, cylindrical second segment; a small third segment with the attachment organ and a cylindrical fourth segment bearing, in the ground pattern, 4+5 sensory setae. It is the third and fourth segments that vary most extensively among cirripedes. This is hardly surprising, since these two segments are in direct contact with the many different types of substrata used in settlement by cyprids of the different species. In acrothoracican cyprids, the most pronounced difference on the 3rd segment is the absence or at least extreme reduction of postaxial seta 3 (ps3). In thoracican cyprids, this seta is a long and conspicuous simple seta, while in rhizocephalan male cyprids it has the form of a long aesthetasc (Walker, 1985; Moyse et al., 1995). Otherwise, the third segment of acrothoracican cyprids has a fairly conventional morphology.
The fourth segment carries four subterminal and five terminal setae in both the Acrothoracica and the Thoracica (Clare & Nott, 1994), and this represents the cirriped ground pattern. Only the Rhizocephala have fewer setae on this segment, and this is probably an apomorphic condition (Høeg & Rybakov, 1996). However, the Acrothoracica do deviate in lacking setulation on any of the terminal setae, whereas the Thoracica have two setulose setae. Grygier (1987a) made a pioneering attempt in homologizing segments and setae in antennules of ascothoracids, facetotectan cypris-y, and cirripede cyprids. Further conclusions must await SEM studies of facetotectan antennules.
Few studies have focused on the cypris thoracopods, despite their importance in the rapid swimming bouts during the pelagic phase of the larva. Cyprids of Lithoglyptes and Trypetesa have a three-segmented endopod and a two-segmented exopod just as in the ground pattern for ascothoracid larvae (Turquier, 1967, present paper). As discussed in detail by Grygier (1987b) and commented on by Glenner & Høeg (1995), this signifies that such a segmentation scheme characterized not only the cypris-like larva of the urthecostracan but also the true cypris of the urcirripede. Our observation on Lithoglyptes also confirms Grygier‘s (1987b) character matrix in that the fusion in cirripedes occurred between endopodal segments 1 and 2. The Facetotecta have a three segmented endopod in the ground pattern, but Schram (1970) and Grygier (1987b) found that some facetotectans have evolved a two segmented state by fusion between endopodal segments 2 and 3. Obviously, the two segmented endopods found in some cirripeds and facetotectans do not represent homologous states and again demonstrates how simple counting of limb articles can lead to erroneous conclusions as elegantly elaborated in Huys & Boxshall (1991).
The apparent lack of flexure between endopod segments one and two in Lithoglyptes cyprids indicates that it was these two segments that fused into one in the evolution of the Thoracica (and Rhizocephala?). In the Thoracica, a faint suture in the first segment of the endopod recalls the plesiomorphic three-segmented condition (Glenner et al., 1995).
All cirripede cyprids carry a single stout and serrated seta on the first exopod segment of all eight thoracopods (Glenner and Høeg, 1995; Walossek et al., 1996 Fig. 14B). These setae are always shorter than the natatory ones on the second segment, and they undoubtedly serve in grooming both the natatory setae and the limb bases. However, they are rarely if ever as large and strongly armed as in the acrothoracican cyprids studied here. Nothing comparable to these grooming setae exists in ascothoracid or cypris-y larvae, so they represent another of the many apomorphies characterizing the cirripede cyprid.
The single seta inserting medially on the 2nd endopod segment is also present in ascothoracid larvae, cypris-y, and in thoracican cypris larvae. But aside from this, it is premature to speculate on the ground pattern of thoracopodal natatory setae in the Thecostraca.
Tagmosis and hindbody
According to Grygier (1983, 1987) and Grygier & Ohtsuka (1995) both the Thecostraca and the Maxillopoda in general have a 5-7-4 tagmosis scheme in the ground pattern. In cyprids of Lithoglyptes the presence of four, short abdominal segments and an elongate telson with unsegmented furcal rami dovetails with this pattern. The apparently 4-segmented abdomen is plesiomorphic compared to all other cirripedes (larval or adult). We found no trace of a 7th thoracomere, which in the thecostracan ground plan carries the penis, unless it forms part of the annulated region we here designate as abdomen. Unpublished SEM micrographs reveal that cyprids of some lepadomorphan Thoracica can have a three-segmented abdomen.
A more or less deeply cleaved telson bearing unsegmented caudal rami constitutes a ground pattern feature in cirripede cypris larvae. It occurs in the Acrothoracica (this study), the Rhizocephala (Walossek et al., 1996) and apparently also in cyprids of lepadid Thoracica (Grygier, 1987b). In contrast, Walker & Lee (1976) and Glenner & Høeg (1995) used SEM to claim that balanomorphan cyprids (Balanus amphitrite, Semibalanus balanoides) have two segmented “caudal rami” inserted directly on the posteriormost end of the thorax and no visible abdomen or telson. We believe that also balanomorph cyprids have unsegmented rami inserted on a telson, but that the cleft have become so deep that the telson can easily be mistaken for “basal ramal segments”. A similar error probably lead Turquier (1967) to describe two segmented “furcal rami” in cyprids of the acrothoracican Trypetesa nassaroides. Support for our claim could come from serial sections revealing that the purported “first ramal segments” are united by a slim medial connection. The urthecostracan undoubtedly had unsegmented caudal rami, since we find this condition in both the Ascothoracida, the Facetotecta, and the outgroup Copepoda (Grygier, 1987b). Obviously, SEM and TEM studies of the hindbody in additional thoracican cyprids may provide characters useful for a phylogenetic analysis.
The data from SEM analysis presented here again highlight the value of larval characters in elucidating thecostracan and cirripede phylogeny (Grygier 1987a,b, 1994, 1995; Jensen et al., 1994; Elfimov, 1995; Moyse et al., 1995; Høeg et al., 1998). The cypris larvae of the Acrothoracica exhibit similarities both with the Ascothoracida and the Facetotecta, and with the two remaining cirripede orders. Although we await a full-fledged phylogenetic analysis, as in Glenner et al. (1995), we will here assume that the similarities with ascothoracid larvae and facetotectan cypris-y represent symplesiomorphies. They include: the “keel in a trough” shape of the lattice organ; three-segmented thoracopodal endopods; a five-segmented hindbody. These plesiomorphies do not alter the fact that the Acrothoracica have a typical cirripede cypris with numerous apomorphies compared to the thecostracan ground pattern, such as: a single pair of frontal (horn) gland pores on the carapace; very similar four-segmented antennules with a homologous attachment organ on the 3rd segment and a 4+5 setation scheme on the fourth segment; paired cement glands terminating on the attachment organ; thoracopods with a stout, serrated seta on the first exopodal segment; abdomen highly shortened. The status of some other characters remains more uncertain or insufficiently analysed: the position of the terminal pore in lattice organs; the presence of a penis rudiment; and the number and special morphology of thoracic natatory setae.
We have in this paper proposed some phylogenetic scenarios based on single characters sets, but we stress that they are meant at this point more as mental exercises than solidly built theories. Yet with studies such as the present one and those of Grygier (1994), Jensen et al. (1994), Moyse et al. (1995), Korn (1995), Elfimov (1995), and Walossek et al. (1996) we are approaching the point where we can enlarge the thecostracan character matrix of Grygier (1987b) and Glenner et al. (1995) with a wealth of new characters from larval morphology.
Clare AS, Nott JA. 1994. Scanning electron microscopy of the fourth antennular segment of Balanus amphitrite amphitrite. J. Mar. Biol. Ass. U.K. 74: 967-970.
Elfimov AS. 1986. Morphology of the carapace in the cypris larva of Heteralepas mystacophora Newman (Cirripedia, Thoracica) [In Russian]. Biol. Morya 1986 (3): 30-34. [English translation in: Soviet J. mar. Biol. 12: 152-156]
Elfimov AS. 1995. Comparative morphology of thoracican larvae: studies on the carapace. Crust. Issues 10: 137-152.
Gibson P, Nott JA. 1971. Concerning the fourth antennular segment of the cypris larva of Balanus balanoides. In: Crisp DJ, ed. Fourth European marine biology Symposium. Cambridge University Press, 227-236
Glenner H. in press. Functional morphology of the cirripede cypris: A comparative approach. In: Thompson MF, Nagabhushanam R, eds. Barnacle Fouling: Ecophysiology and Control Technology. Washington :A.I.B.S.
Glenner H,Grygier MJ, Høeg JT, Jensen PG, Schram FR. 1995. Cladistic analysis of the Cirripedia Thoracica (Crustacea: Thecostraca). Zool. J. Linn. Soc. 114: 365-404.
Glenner H, Høeg JT. 1995. Scanning electron microscopy of cypris larvae of Balanus amphitrite amphitrite (Cirripedia: Thoracica: Balanomorpha). J. Crustacean Biol. 15: 523-536.
Glenner H, Høeg JT, Klysner A, Brodin Larsen B. 1989. Cypris ultrastructure, metamorphosis and sex in seven families of parasitic barnacles (Crustacea: Cirripedia: Rhizocephala). Acta Zool. (Stockholm) 70: 229-242.
Gotelli NJ, Spivey HR. 1992. Male parasitism and intrasexual competition in a burrowing barnacle. Oecologia 91: 474-480.
Grygier MJ. 1983. Ascothoracida and the unity of the Maxillopoda. Crust. Issues 1: 75-103.
Grygier MJ. 1987a. Nauplii, antennular ontogeny, and the position of the Ascothoracida within the Maxillopoda. J. Crustacean Biol. 7: 87-104.
Grygier MJ. 1987b. New records, external and internal anatomy, and systematic position of Hansen‘s Y-larvae (Crustacea: Maxillopoda: Facetotecta). Sarsia 72: 261-278.
Grygier, M.J. 1988. Larval and juvenile Ascothoracia (Crustacea) from the plankton. Publ. Seto Mar. Biol. Lab. 33: 163-172
Grygier MJ. 1994. Developmental patterns and updated hypotheses of homology in the antennules of Thecostracan nauplius larvae (Crustacea). Acta Zool. (Stockholm) 75: 219-234.
Grygier MJ. 1995. An unusual barnacle nauplius illustrating several hitherto unappreciated features useful in cirripede systematics. Crust. Issues 10: 123-136.
Grygier MJ, Newman WA. 1985. Motility and calcareous parts in extant and fossil Acrothoracica (Crustacea: Cirripedia), based primarily upon new species burrowing in the deep-sea coral Enallopsammia. Trans. San Diego Soc. Nat. Hist. 21: 1-22.
Grygier MJ, Ohtsuka S. 1995. New species of Synagoga (Crustacea: Ascothoracida) from plankton off Okinawa, Japan, with a SEM study of the carapace. Publ. Seto Mar. Biol. Lab. 36: 393-311.
Høeg JT. 1985. Cypris settlement, kentrogon formation and host invasion in the parasitic barnacle Lernaeodiscus porcellanae (Müller) (Crustacea: Cirripedia: Rhizocephala). Acta Zool. (Stockholm) 66: 1-45.
Høeg JT. 1992. The phylogenetic position of the Rhizocephala: Are they truly barnacles?. Acta Zool. (Stockholm) 73: 323-326.
Høeg JT. 1995. Sex and the single cirripede: a phylogenetic perspective. Crust. Issues 10: 195-206.
Høeg JT, Hosfeld B, Jensen PG. 1998. TEM studies of lattice organs of cirriprede cypris larvae (Crustacea, Thecostraca, Cirripedia). Zoomorphology 118: 195-205.
Høeg JT, Rybakov AV. 1996. Cypris ultrastructure in Arcturosaccus kussakini (Rhizocephala) and the homology of setae on the fourth antennular segment in rhizocephalan and thoracican cyprids. Zool. Anz . 234: 241-251.
Hosfeld B, Høeg JT, Jensen PG. 1998. Ultrastructure of lattice organs in facetotectan cypris-y: implications for thecostracan phylogeny. 4th International Crustacean Congress, Amsterdam, July 20-24, 1998. Abstract volume.
Huys R, Boxshall GA. 1991. Copepod Evolution . London: Ray Society.
Itô T, Grygier MJ. 1990. Description and complete larval development of a new species of Baccalaureus (Crustacea: Ascothoracida) parasitic in a zoanthid from Tanabe Bay, Honshu, Japan. Zool. Sci. 7: 485-515.
Jensen PG, 1993. Ultrastructure and phylogenetic significance of ‘lattice organs‘ in thecostracan larvae. Amer. Zool. 33 (5): 6A
Jensen PG, Moyse J, Høeg JT, Al-Yahya H. 1994. Comparative SEM studies of lattice organs: Putative sensory structures on the carapace of larvae from Ascothoracida and Cirripedia (Crustacea Maxillopoda Thecostraca). Acta Zool. (Stockholm) 75: 125-142.
Klepal W, Nemeschkal HL. 1995. Cuticular structures in the males of Scalpellidae (Cirripedia Thoracica): A character analysis. Crust. Issues 10: 179-194.
Kolbasov GA. 1996. The significance of symbiosis in the evolution of sessile barnacles (Cirripedia Balanoidea). Arthropoda Selecta 5(1/2): 3-16.
Korn OM. 1995. Naupliar evidence for cirripede taxonomy and phylogeny. Crust. Issues 10: 87-121.
Kühnert L. 1934. Beitrag zur Entwicklungsgeschichte von Alcippe lampas Hancock. Z. Morph. Ökol. Tiere 29: 45-78.
Moyse J, Jensen PG, Høeg JT, Al-Yahya H. 1995. Attachment organs in cypris larvae: using scanning electron microscopy. Crust. Issues 10: 153-178.
Newman WA. 1971. A deep-sea burrowing barnacle (Cirripedia: Acrothoracica) from Bermuda. J. Zool., London 165: 423-429.
Newman WA. 1974. Two new deep-sea Cirripedia (Ascothoracica and Acrothoracica) from the Atlantic. J. Mar. Biol. Ass. U.K. 54: 437-456.
Newman WA. 1982. Cirripedia. In: L. G. Abele LG, ed. The Biology of the Crustacea 1. Systematics, the Fossil Record, and Biogeography. London, New York: Academic Press, 197-221
Newman WA. 1987. Evolution of cirripedes and their major groups. Crust. Issues 5: 3-42.
Newman WA. 1996. Sous-Classe des Cirripèdes (Cirripedia Burmeister, 1834). Super-ordres des Thoraciques et des Acrothoraciques (Thoracica Darwin, 1854 - Acrothoracica Gruvel, 1905). In: Forest J, ed. Traité de Zoologie, Tome VII Fasc. II, Crustacés: Generalites (suite) et Systématique (1. parti). Paris: Masson: 453-540.
Nott J, Foster B. 1969. On the structure of the antennular attachment organ of the cypris larva of Balanus balanoides (L. ). Phil. Trans. R. Soc. London 256B: 115-134.
Schram TA. 1970. Marine biological investigations in the Bahamas 14. Cypris y, a later developmental stage of nauplius y Hansen. Sarsia 44: 9-24.
Spears T, Abele LG, Applegate MA. 1994. A phylogenetic study of cirripeds and their relatives (Crustacea Thecostraca). J. Crustacean Biol. 14: 641-656.
Tomlinson JT. 1969. The burrowing barnacles (Cirripedia: order Acrothoracica). Bull. U.S. Natl. Mus. 296: 1-162.
Turquier Y. 1967. Le dévelopment larvaire de Trypetesa nassaroides Turquier (Cirripède Acrothoracique) et ses rapports avec celle des autres Cirripèdes. Arch. Zool. Exp. Gén. 108: 33-47.
Turquier Y. 1970. Recherches sur la biologie des Cirripèdes Acrothoraciques. III. La métamorphose des cypris femelles de Trypetesa lampas (Hancock) et de Trypetesa nassaroides Turquier. Arch. Zool. Exp. Gén. 111: 573-627.
Turquier Y. 1971. Recherches sur la biologie des Cirripèdes Acrothoraciques. IV. La métamorphose des cypris mâles de Trypetesa nassaroides Turquier et de Trypetesa lampas (Hancock). Arch. Zool. Exp. Gén. 112: 301-348.
Turquier Y. 1972. Contribution à la connaisance des Cirripèdes Acrothoraciques. Arch. Zool. Exp. Gén. 113: 499-551.
Turquier Y. 1985. Cirripèdes Acrothoraciques des côtes occidentales de la Mediterranée et de l‘Afrique du Nord. II. Weltneria zibrowii n.sp. Bull. Soc. Zool. Fr. 110: 169-189.
Walker G, 1985. The cypris larvae of Sacculina carcini Thompson (Crustacea: Cirripedia: Rhizocephala). J. Exp. Mar. Biol. Ecol. 93: 131-145.
Walker G, Lee V. 1976. Surface structure and sense organs of the cypris larva of Balanus balanoides by scanning and transmission electron microscopy. J. Zool. London 178: 161-172.
Walker G, Yule AB, Nott JA. 1987. Structure and function of balanomorph larvae. Crust. Issues 5: 307-328.
Walley LJ. 1969. Studies on the larval structure and metamorphosis of Balanus balanoides (L. ). Phil. Trans. R. Soc. Lond. 256B: 237-280.
Walossek D, Høeg JT, Shirley TC. 1996. Larval development of the rhizocephalan cirripede Briarosaccus tenellus (Maxillopoda: Thecostraca) reared in the laboratory: A scanning electron microscopy study. Hydrobiologia 328: 9-47.
Wells HW, Tomlinson JT. 1966. A new burrowing barnacle from the Western Atlantic. Quart J. Florida Acad. Sci. 29: 27-37.
JTH and GAK express their gratitude to H. Glenner, J. Olesen and T. Schiötte for help with SEM preparation and operation. Grants no. 11-9652 and no. 9701589 from the Danish Natural Science Research Council enabled JTH to finance GAK visits to Denmark and are gratefully acknowledged. GAK thanks the Russian Foundation for Basic Research (grant N 97-04-49803) for supporting his work in Moscow State University. GAK also acknowledges Dr. D.L. Ivanov (the chief of the mollusc collection of the Zoological Museum of Moscow State University) and R.V. Egorov for an opportunity to examine the collections of the Zoological Museum and help in the species determination of mollusc shells. ASE and GAK are finally indebted to G.N. Davidovich and I.A. Bogdanov of the Laboratory of Electron Microscopy of Biological Faculty of Moscow State University for help with SEM investigations.