Showing papers in "Invertebrate Biology in 1999"

A Molecular Test of Platyhelminth Phylogeny: Inferences from Partial 28S rDNA Sequences

Abstract: Nucleotide sequences of the region extending from the D3 to the D6 expansion segments of the 28S rDNA gene were used to reconstruct evolutionary relationships within the Platyhelminthes. Neighbor-joining and parsimony analyses of representatives of most major platyhelminth taxa revealed a basal Catenulida, a monophyletic Acoelomorpha, a sister-group relationship of Macrostomorpha and Polycladida (59% of bootstrap replications), and monophyletic Tricladida. We found no evidence of a taxon Seriata (p=.0001); however, the paraphyletic status of the Proseriata needs further investigation. Although Neodermata appeared as a monophyletic group, Monogenea was paraphyletic (p<.0001), thus not supporting the taxon Cercomeromorpha. Monopisthocotylean Monogenea was the most basal taxon among the Neodermata, and Udonella sp. consistently grouped with the Monopisthocotylea. Other relationships within the Neodermata showed Trematoda as a monophyletic taxon and, among its members, Rugogaster hydrolagi was identified as an aspidogastrean. The taxa Cestodaria (Amphilinidae and Gyrocotylidae) and Eucestoda were both supported; however, constraining these 2 clades into a monophyletic Cestoda resulted in a significantly longer tree (p=.0303). We tentatively conclude that the immediate sister group of Neodermata is Fecampiida. But because of contradictory results of the constraint analyses, we cannot exclude the possibility that the sister group to Neodermata is a taxon containing many neoophoran turbellarians (e.g., Rhabdocoela and Tricladida). Additional key words: Acoelomorpha, Neodermata, Udonellida, molecular systematics Although the Platyhelminthes, and especially the order Acoela, hold a key position in several theories of metazoan origins (Hadzi 1963; Steinbock 1963; Salvini-Plawen 1978; Ehlers 1985; Ax 1987; Nielsen 1987), their within-group relationships are poorly understood. Attempts at classifying the platyhelminths have produced a wealth of ultrastructural information, but none of these studies have resulted in a definitive phylogenetic tree (Smith et al. 1986). However, two major points regarding platyhelminth systematics have emerged from these ultrastructural studies. First, Turbellaria may be a paraphyletic group, and the term should probably be used with caution (Ax 1987). Secondly, 3 clearly defined monophyletic lineages have been identified within the phylum (Fig. 1): the Catenulida; the Acoelomorpha, consisting of the orders Acoela and Nemertodermatida; and the Rhabditophora, which comprises the remaining orders and the parasitic Neodermata (Ehlers 1985, 1986). To determine a Author for correspondence. E-mail: mkll @cisunix.unh.edu relationships among these 3 groups, however, is difficult because of a lack of convincing synapomorphies. Proposed characters such as "frontal organ/frontal glands" or "epidermal replacement" are either not homologous across the taxa (Klauser et al. 1986; Smith & Tyler 1986) or have only an extremely low probability of homology (Smith et al. 1986), respectively. Therefore, monophyly for the platyhelminths, intuitive though it may be, will be difficult to test using traditional morphological characters. In the last decade, molecular phylogenies at any taxonomic level have gained great popularity. Nucleotide sequences provide an alternative data set for the resolution of evolutionary relationships in cases where morphologists may have reached an impasse regarding a specific phylogeny, or where the number of species and structural characters needed to reliably resolve a phylogeny may be prohibitively high. Molecular systematics has attempted to resolve just such issues by providing large and novel data sets (Turbeville et al. 1991; Turbeville et al. 1992). Some of the most popThis content downloaded from 207.46.13.164 on Sun, 26 Jun 2016 07:38:12 UTC All use subject to http://about.jstor.org/terms Molecular platyhelminth phylogeny MonoI,uI pI lal Monopisthocotylea genea Neo-(incl. Udonella sp.) dermat; L Digenea [-EAspidogastrea Trematoda (incl. R. hydrolagi) Dalyellioida (incl. Fecampiida) RhabdoKalyptorhynchia coela Tricladida Seriata Proseriata

An Introduced Sabellid Polychaete Pest Infesting Cultured Abalones and Its Potential Spread to Other California Gastropods

Abstract: In 1993 an unusual sabellid polychaete was brought to our attention. It grossly deformed the shells of cultured abalones in some of the California abalone mariculture facilities. This worm is not native to California and was subsequently found in rocky intertidal and subtidal habitats in southern Africa, where it had not previously been recognized. The worm is hermaphroditic and has benthic larvae that are competent to settle within 12 h and soon secrete a mucous sheath. Development of the tentacular crown occurs within a week and generation time can be short, about one month. The worm has a unique association with host gastropods. Unlike all other known shell-fouling organisms, the sabellid routinely settles inside the aperture at the growing edge of the shell. The host responds by secreting a layer of nacre over the mucous sheath to form a tube enclosing the worm, whose crown of tentacles extends through the opening of the tube to the outer surface of the shell. Heavy infestations cause the cessation of linear growth of the host as prismatic shell deposition cannot be resumed after repeated settlement of larvae. The sabellid is not very host specific; many other California native gastropods are readily infested. Bivalves do not appear to be susceptible. Efforts to find a native California predator of the adult worms were not successful. The sabellid has caused great economic damage to some facilities commercially culturing abalones. An established population of this worm has been detected in California, and further risk of establishment and spread of this worm is great. Its unique biology suggests that it may be a useful experimental probe for studies of molluscan shell deposition and may also serve to reveal how molluscs defend themselves against organisms attempting to settle in and foul their apertures. Additional key words: molluscan mantle, shell deposition, fouling, South Africa In October 1993, growers at some commercial abalone mariculture facilities in southern California brought an interesting sabellid polychaete to our attention. The worm infested shells of cultured red abalones, Haliotis rufescens. Heavily infested abalones ceased growth. Their shells exhibited an abnormal, beaked shape and often lacked respiratory holes. The worms appeared to bore in the shell. Because this worm had not previously been reported from California and only a few species of shell-boring sabellid in the genus Caobangia were known (Jones 1969), it was of obvious systematic and biological interest. It has been described as a new genus and species, Terebrasabella heterouncinata FITZHUGH & ROUSE 1999. It is evidently native to at least South Africa, as recognized by Ruck & Cook (1998). Several types of polychaetes infest mollusc shells. a Author for correspondence. E-mail: kuris@lifesci.ucsb.edu Spirorbids are sometimes abundant epibionts, often settling on the outer surfaces of molluscan shells (Gee & Knight-Jones 1962; Zuschin & Piller 1997). They also are sometimes very abundant on the shells of cultured abalones in California. Some serpulids are also epibionts. Spionid polychaetes, particularly species of Polydora and Boccardia, known as mudworms, infest molluscan shells. They settle on the outer surface and bore into the shell, often forming a U-shaped or flasklike burrow. The mechanisms by which spionids bore have received considerable attention (Dorsett 1961; Haigler 1969; Zottoli & Carriker 1974; Sato-Okoshi & Okoshi 1993), and these worms can cause damage to both cultured and wild populations of molluscs by weakening shells and increasing their susceptibility to predation, or by loss of energy expended in shell repair (Kent 1979; Bergman et al. 1982; Kojima & Imajima 1982; Mori et al. 1985; Sato-Okoshi and Nomura 1990; Ambariyanto & Seed 1991; Wargo & Ford This content downloaded from 207.46.13.129 on Sun, 26 Jun 2016 07:01:20 UTC All use subject to http://about.jstor.org/terms

Endobionts of the Coral Reef Sponge Theonella swinhoei (Porifera, Demospongiae)

Abstract: Endobionts associated with the coral reef sponge Theonella swinhoei (Lithistida, Demospongiae), from the Red Sea and Indian Ocean, were characterized through histological and ultrastructural investigations. Prokaryotes included unicellular heterotrophic bacteria (mean concentration, 20% of the sponge tissue), unicellular cyanobacteria (Aphanocapsa feldmanni, 15% of the sponge tissue), and non-photosynthetic filamentous forms, probably in the Beggiatoa-group (sulfur bacteria, family Beggiatoaceae, 40% of the sponge tissue). Evidence of ingestion of small pieces of filamentous bacteria by sponge cells was observed. The worm Haplosyllis spongicola (Syllidae, Polychaeta) inhabited the canals of the sponge aquiferous system (mean density, 73 worms/cm3). The abundance and distribution of the endobionts living in T. swinhoei from different localities are presented and discussed in comparison with those of other sponge species reported in literature. Additional key words: polychaetes, filamentous bacteria, cyanobacteria Sponges present an astonishing landscape of associations with various kinds of autotrophic and heterotrophic organisms (Pearse 1932; Pansini 1970; Ilan et al. 1994). The various parts of a sponge, especially the endosome, contain large populations of extracellular bacteria (Vacelet 1975; Vacelet & Donadey 1977), which may greatly differ from the bacterial population in the water surrounding the same sponge (Wilkinson 1978b). The presence of cyanobacteria or other photosynthetic organisms is less frequent. Even though the autotrophic organisms inhabit the more superficial layers in order to fulfill their metabolic needs, some peculiar localizations inside the skeletal fibers (Riitzler 1990) and along the siliceous spicule bundles (Gaino & Sara 1994) have been observed. In this last condition, siliceous spicules, acting as a light-conducting system (Cattaneo-Vietti et al. 1996), enhance photosynthesis. Since the first observations of Feldmann (1933), many ultrastructural studies have stressed the presence of cyanobacteria in sponges of both Mediterranean and tropical origin (Sara 1971; Vacelet 1971; Vacelet & Donadey 1977; Wilkinson 1978a,c; Rutzler 1990). In some cases symbionts are so tightly associated with the sponge tissues as to constitute a "symbiocortex" (Simpson 1984, pp. 114-123). Cyanobacteria protect a Author for correspondence. E-mail: zoologia@unige.it sponges from excessive solar radiation (Sara & Vacelet 1973; Wilkinson 1980) and variations in the photosynthetic symbiont concentration may effect changes in the color of the sponge surface (Gaino et al. 1977). Direct feeding on cyanobacteria has been frequently suggested but observed only for unicellular types (Sara 1966; Wilkinson 1980; Berthold et al. 1982; Rutzler 1990). The translocation of glycerol, fixed nitrogen, and other photosynthetically fixed nutrients represents a trophic advantage for the sponge host (Wilkinson 1979; Wilkinson & Fay 1979), mainly in coral reefs and other tropical environments where the low levels of available nutrients favor these associations (Wilkinson 1987). So far, only two species of unicellular cyanobacteria have been frequently described in association with sponges: Aphanocapsa feldmanni and A. raspaigellae (Sara 1966; Gaino et al. 1976; Wilkinson 1980; Rutzler 1990). A few reports treat filamentous multicellular cyanobacteria, differing in dimension and shape, most commonly Oscillatoria spongeliae (Sara 1966) and Phormidium spongeliae (Wilkinson 1980). Nonetheless, Bewley et al. (1996), in ultrastructural observations, reported the lack of a photosynthetic system in some prokaryotic filaments, previously attributed to the above mentioned cyanobacteria. In addition, sponges serve as hosts to a number of other heterotrophic organisms, providing food and shelter to species from many taxa. Polychaetes have This content downloaded from 157.55.39.153 on Mon, 19 Sep 2016 04:41:51 UTC All use subject to http://about.jstor.org/terms Magnino, Sara, Lancioni, & Gaino been commonly observed in sponges (Pearse 1950; Pawlik 1983; Koukouras et al. 1996; Martin 1996; Rutzler, 1997). In particular, the polychaete Haplosyllis spongicola (GRUBE 1855), family Syllidae, is an abundant inhabitant of sponges. Fauchald & Jumars (1979) proposed that this worm grazes on the sponge surface. A more intimate relationship with its host has been directly observed by Tsurumi & Reiswig (1997), who reported that H. spongicola feeds on the tissues surrounding the sponge water canals, within which it lives. It is often difficult to establish the correct relationship between sponges and their endobionts. The sponge Theonella swinhoei GRAY 1868 (order Lithistida) is a remarkable component of the sponge populations in coral reef communities. Heterospecific associations, representing many types of adaptive interactions, are very common in this peculiar environment (Wilkinson 1987; Castro 1988). Previous studies on T. swinhoei from the Red Sea showed the unicellular symbionts A. feldmanni associated with sponge tissues (Wilkinson 1979, 1980; Bewley et al. 1996). Here we document that specimens of T. swinhoei, from both Red Sea and Indian Ocean coral reefs, constitute a further example of how sponges may represent a suitable micro-environment for the development of various kinds of autotrophic and heterotrophic organisms, which inhabit different parts of the sponge body.

High Fecundity and Low Fertility in Parthenogenetic Planarians

Abstract: In hermaphrodites, the cost of sexuality can favor the spread of parthenogenesis in two ways. First, it can promote higher female fecundity in parthenogens that have reduced allocation to the male function. Second, if parthenogens have retained a fertile male function, they can spread genes for parthenogenesis into a coexisting sexual population. We present evidence for both effects in a natural population of the flatworm Schmidtea polychroa. Parthenogens, which have a reduced male function, had 42% higher female fecundity than coexisting sexuals. New, presumably parthenogenetic, triploids arose out of the diploid sexual population at a frequency of 1.3%, probably as a result of gene flow from parthenogens to sexuals. However, we could also identify a strong compensating fitness benefit for sexuals: they had substantially higher female fertility than coexisting parthenogens, both in terms of cocoon fertility (93% and 74% respectively) and offspring per fertile cocoon (3.6 and 2.8 respectively). Additional key words: evolution of sex, sex allocation, Platyhelminthes, Dugesia, Schmidtea Evolutionary theory predicts that strong counteracting selective forces act on sexual and parthenogenetic reproduction. On the one hand, it is widely accepted that sex pays the high "cost of male allocation" (Maynard Smith 1978) or "cost of meiosis" (Williams 1975). On the other hand, benefits of sexual reproduction have been postulated such as advantages in hostparasite coevolution, reduced accumulation of mutations, reduced competition among sexuals, and the opportunity to repair physical DNA damage (Bell 1982; Vrijenhoek 1984; Stearns 1987; Kondrashov 1988; Michod & Levin 1988; Weeks 1993). Some predictions of these theories have been the focus of experimental studies (Lively 1987; Mogie & Ford 1988; Michaels & Bazzaz 1989; Hamilton et al. 1990; Lively et al. 1990; Moritz et al. 1991; Vrijenhoek 1994), but relatively little effort has been put into quantifying the costs and benefits of sex. The only available field estimate of the benefit of sex is the 1.43-fold higher fitness of sexually derived offspring that was found in an experimental study of the grass Anthoxanthum odoratum (Kelley et al. 1988). Estimating the cost of sex in hermaphrodites is not as easy as in outcrossing gonochoristic species, and has been the subject of several theoretical studies (Charlesworth 1980; Lloyd 1988; Joshi & Moody a Author for correspondence. Present address: Institut fiir Spezielle Zoologie, University of Muenster, Huefferstrasse 1, D-48149 Muenster, Germany. E-mail: michiels @ uni-muenster.de 1995). Briefly summarized, it depends on the sex allocation of the coexisting sexual and parthenogenetic hermaphrodites, as well as on the ability of the male function of the parthenogens to generate parthenogenetic offspring in sexual partners. If the latter is not possible, the male function of a parthenogen is effectively "sterile" for the parthenogenetic sub-population, as it makes no contribution to the next generation of parthenogens. Selfing rate and inbreeding depression are also important, but not relevant for our model system and are therefore not mentioned here. Let us first consider a newly arisen parthenogenetic lineage with the same sex allocation as its sexual ancestor, but with a sterile male function (e.g., parthenogens are polyploid and so are their sperm). Such individuals will produce the same number of maternal offspring and make the same genetic contribution to the next generation as sexuals. Hence, the cost of sex is zero and additional changes are needed to offer an advantage to parthenogenetic hermaphrodites (see also Mogie 1996). On the one hand, parthenogens may reduce male allocation rparth and increase female allocation (1-rparth). Only when rse=0.5 in the sexual ancestor, and by complete elimination of the male function in parthenogens, can the latter obtain a 2-fold advantage. In all other cases (rsex O), the cost of sex [1/(1-rsex)] will be less than 2-fold. On the other hand, parthenogenetic hermaphrodites may generate new parthenogenetic lineages by fatherThis content downloaded from 157.55.39.45 on Thu, 01 Sep 2016 04:44:16 UTC All use subject to http://about.jstor.org/terms Weinzierl, Schmidt, & Michiels ing eggs of sexual partners. Assuming a simple genetic system, and all else being equal between sexuals and parthenogens, this will result in a (maximum) 3/2 advantage for parthenogens (Charlesworth 1980). This advantage will be lower when some partners of parthenogens are not sexual, but parthenogenetic as in sperm-dependent parthenogens (Beukeboom & Vrijenhoek 1998). Loss of eggs and sperm from sexuals to parthenogenetic partners should also be included in the cost of sex. Hence, in order to reveal the cost of sex in a natural population of hermaphrodites one has to consider (1) reallocation of resources to the female function in parthenogens and (2) the production of parthenogenetic offspring through gene flow from parthenogens to sexuals. Many examples of reduced male function in parthenogenetic hermaphrodites are known (Christensen et al. 1978; Nogler 1984; 0 Foighil & Eernisse 1988; 0 Foighil & Thiriot-Quievereux 1991; Weinzierl et al. 1998), but the reduction has never been shown to lead to a substantial increase in female reproductive success. This may be explained by the fact that male allocation is low in some sexual hermaphrodites (Bell 1984a,b). Yet, it is clearly high in others (e.g., Rameau & Gouyon 1991). Gene flow from parthenogens to sexuals is possible in a number of hermaphrodites (Suomalainen et al. 1987; Asker & Jerling 1992) and can result in new parthenogenetic lineages (Menken et al. 1995). The hermaphroditic freshwater planarian Schmidtea polychroa BALL (=Dugesia polychroa (SCHMIDT); Tricladida, Paludicola) consists of a diploid sexual and several polyploid parthenogenetic biotypes (Benazzi 1957). Individuals of both modes produce cocoons that contain 1-10 eggs embedded in a common yolk mass. Parthenogenetic reproduction is pseudogamous, i.e., egg development is stimulated by a sperm cell that does not contribute genetically (Benazzi 1950; Benazzi & Benazzi Lentati 1976; Beukeboom & Vrijenhoek 1998). Despite being polyploid, parthenogenetic animals have functional, haploid sperm (Benazzi Lentati 1970), but, like sexuals, always need allosperm from a sexual or parthenogenetic partner in order to produce maternal offspring (Benazzi & Benazzi Lentati 1992). Parthenogens mate frequently, with each other and with sexuals (Peters et al. 1996; M. Storhas, pers. comm.). In S. polychroa, as in other planarians (Hyman 1951), testes in sexual individuals are numerous and occupy significant portions of the body. Parthenogens of S. polychroa, in contrast, have very few testes (Weinzierl et al. 1998). Parthenogenetic worms also appear to have a lower mating rate than sexuals (Weinzierl et al. 1998), which again saves resources that can be reallocated to the female function. We tested whether parthenogens of S. polychroa have higher female fecundity than their coexisting sexual conspecifics. Flexible reallocation is likely in planarians, because they are known to adjust cocoon production in response to fluctuating resource availability (Reynoldson 1964, 1977; Reynoldson & Young 1965; Boddington & Mettrick 1977; Calow & Woolhead 1977). By fertilizing sexual eggs, parthenogens of S. polychroa can "inject" genes for parthenogenesis into the coexisting sexual population, and this can generate new parthenogenetic lineages in laboratory crosses (Benazzi Lentati 1966). We investigated whether sexual individuals taken directly from our study population produced parthenogens among their offspring. One balancing selective force favoring sexual reproduction is female fertility, which is known to be reduced in a number of parthenogenetic insects (Lamb & Willey 1979). We compared fertility of cocoons produced by sexual and parthenogenetic worms.

Endostyle-like Features of the Dorsal Epibranchial Ridge of an Enteropneust and the Hypothesis of Dorsal-Ventral Axis Inversion in Chordates

Abstract: Classical anatomical investigations of the spengelid enteropneust Schizocardium brasiliense suggested that the hypobranchial ridge in the ventral midline of the pharynx is a homolog of the chordate endostyle. A re-investigation of pharyngeal anatomy and histology of S. brasiliense does not support this homology. Instead, the dorsal epibranchial ridge of the pharynx of S. brasiliense provides anatomical and histological correspondences with the ventral endostyle of chordates. The potential homology of a dorsal structure in Enteropneusta with a ventral one in Chordata is consistent with a recent evolutionary model for dorsal-ventral axis inversion in the evolution of chordates. Accepting this homology requires rejecting homology between the enteropneust stomochord and neurocord (collar cord) and the chordate notochord and nerve cord, respectively, but suggests a homology between the enteropneust ventral nerve cord and the chordate neural tube. We propose that functional inversion of the dorsal-ventral body axis occurred in the vertebrate clade and that the ancestor of the vertebrates lacked dorsalventral axis preference, as illustrated by the functional biology of protochordates. Moreover, body axis shifts may have occurred elsewhere in deuterostomes, particularly in the evolution of Echinodermata and Ascidiacea. Additional key words: Hemichordata, Vertebrata, dorsal-ventral axis inversion, evolution, filter feeding An endostyle is an organ situated in the pharyngeal midline of protochordates and ammocoete larvae of lampreys, all of which filter feed using a pharynx and gill slits (Young 1981; Ruppert & Barnes 1994; Bone 1998). All endostyles are associated with at least one pair of ciliated peripharyngeal bands that originates at the anterior end of the endostyle and encircles the pharynx (Delage & Herouard 1898; Bone 1998). As cilia (most protochordates) or muscles (salps, ammocoetes) pump water through the pharynx, the endostyle secretes mucus on which suspended particles are trapped and conveyed into the midgut. Thus, the endostyle is a crucial element in the feeding apparatus of protochordates and, primitively, of vertebrates (Olsson 1963; Godeaux 1989). Because it is a constant and unique feature of the chordate filter-feeding pharynx and is composed of similar parts in all the constituent groups, the endostyle is widely regarded as an autapomorphy of the Chordata (Jefferies 1986; Nielsen 1995; Ruppert 1997a). a Author for correspondence. E-mail: ruppere@clemson.edu Among protochordates, the endostyle is a midventral ciliated gutter or groove that is co-extensive with the pharynx, but in the ammocoete larva, the endostyle is a set of several parallel and coiled tubes, submerged below the pharyngeal lining (Leach 1939). This subpharyngeal endostyle communicates with the pharynx lumen via a single pore. Although the anatomy of the ammocoete endostyle differs strikingly from that of the protochordates, the two designs are considered to be homologous because of similarities in development, histology, and biochemical properties of the secretory cells (Gorbman & Creaser 1942; Thomas 1956; Olsson 1963; Barrington & Thorpe 1965; Ericson et al. 1985; Fredriksson et al. 1985). In particular, the selective binding of iodine to tyrosine residues in the production of mucus by particular cells of the endostyle has been demonstrated for the endostyles of all protochordates (except Thaliacea) and in ammocoetes (Gorbman & Creaser 1942; Thomas 1956; Barrington 1957, 1958; Dunn 1974; Fredriksson et al. 1984, 1985). Although iodine incorporation is constantly associated with some of the endostylar secretory products, This content downloaded from 207.46.13.159 on Sat, 22 Oct 2016 04:29:43 UTC All use subject to http://about.jstor.org/terms Enteropneust epibranchial ridge and chordate endostyle the morphology of the secretory product appears to differ among taxa. In the Ascidiacea, the group in which the endostylar secretion has been studied best, the endostyle constructs a mucous net that is drawn over the lining of the pharynx by frontal cilia. The net comprises threads forming a uniform orthogonal mesh (Flood & Fiala-Medioni 1979, 1981). Each mirrorimage half of the ascidian endostyle typically consists of 8 zones of distinctive cells, most or all of which contribute a component to the final net. Models of mucus-net construction by the endostyle have been proposed (Holley 1986; Godeaux 1989). The endostyle of Thaliacea (Godeaux 1981; Godeaux et al. 1998) is histologically similar to that of ascidians and it also produces a mucous net (Madin & Deibel 1998). The appendicularian endostyle is composed of 5 or 6 mirror-image zones of cells (Olsson 1965), which are thought to cast an irregular mucous net onto the luminal wall of the pharynx (Deibel 1986, 1998), but details of net morphology require further investigation. Cephalochordates of the genus Branchiostoma have an endostyle that bears 6 zones of cells (Barrington 1958; Ericson et al. 1985; Ruppert 1997b). The cephalochordate endostyle is believed to cast a net onto the lining of the pharynx (Ruppert 1997b), but a detailed description of this "net" is lacking. According to Newth's (1930) study of feeding of an ammocoete, the endostylar secretion is released through the pore into the pharynx as strings of mucus, which are transported by the ciliated peripharyngeal bands. The strings form a funnel, its mouth directly anteriorly, that is twisted posteriorly into a cord of mucus and food and then transported into the stomach. Because endostylar product may be fabricated into either strings or nets (and perhaps other forms), depending on taxon, and because an "outgroup" with an endostyle has not yet been identified, it is impossible to infer which form of secretion constitutes the primitive arrangement. On the weak basis of simplicity vs. complexity, it is tempting to consider that the production of mucous strings, transported on or in ciliary bands, preceded the evolution of a mucous net, but evidence with which to test this hypothesis remains to

The Choanosome of Hexactinellid Sponges

Abstract: Three-dimensional images of the choanosome of hexactinellid sponges can help in explaining how these sponges mix cellular and syncytial tissues in a fine cobweb-like network. Plastic replicas of the water canals show that Rhabdocalyptus dawsoni has large, highly branched incurrent canals, and equally large, although less branched, excurrent canals that are studded with 60 (im-long flagellated chambers. Freeze-fractured, fixed specimens show the syncytial tissue, known as the trabecular reticulum, envelops cells in a thin collagenous mesohyl at the flagellated chambers in both R. dawsoni and Aphrocallistes vastus. Groups of archaeocytes are especially abundant in the mesohyl of R. dawsoni during summer months. A branch of the trabecular reticulum, the secondary reticulum, surrounds the collars in the flagellated chambers, effectively forcing water to be drawn through the collar microvilli. Another branch, the inner membrane, occupies up to 10% of flagellated chambers of sponges during all months of the year, but is especially prevalent in specimens which have been kept in sea water aquaria for several weeks. This three-dimensional view of hexactinellid tissues reinforces the conclusion that hexactinellid sponges be separated from other sponges at the subphylum level. Additional key words: Hexactinellida, Porifera, syncytium, sponge Sponge morphology has been studied for well over a century. These animals are considered to be simple, radially symmetric or asymmetric, diploblastic metazoans that carry out all the basic functions of ingestion and egestion, growth and reproduction with a very small complement of cell types. The Hexactinellida, however, differ fundamentally from other sponges in having a syncytial tissue (the trabecular reticulum) rather than cells, as their major tissue component (Reiswig 1979; Mackie & Singla 1983; Leys 1995). The uniqueness of the construction of the Hexactinellida has been recognized since these animals were first examined in the late 1800s. Bidder (1929) first proposed separating the Porifera into two phyla based partly on the difference in tissue organization between cellular and syncytial sponges. Later, Reid (1963) argued that this difference should be recognized at the subphylum level, but the lack of ultrastructural proof of syncytial tissues prevented sponge workers from unanimously accepting the proposal (Bergquist 1978). Because hexactinellids are predominantly deep water sponges, specimens were usually obtained by dredging, and consequently were in poor condition. The discovery of shallow-water populations accessible by a Present address: Department of Zoology, University of Queensland, Brisbane, QLD 4072, Australia. SCUBA in the late 1970s allowed the first ultrastructural examination of tissues and confirmed that the major tissue component in the sponge constitutes a single giant syncytium (Reiswig 1979; Mackie & Singla 1983). Advances in our understanding of the physiology (Perez 1996; Wyeth et al. 1996; Leys & Mackie 1997), development (Boury-Esnault & Vacelet 1994; BouryEsnault et al. 1999), and molecular biology (Koziol et al. 1997; Kruse et al. 1998) of hexactinellids have confirmed the unusual status of this group within the phylum Porifera. However, recent invertebrate texts have yet to acknowledge a proposal for subdivision of sponges into two subphyla (Reiswig & Mackie 1983). Because hexactinellid tissue is unusually difficult to preserve well for either transmission or scanning electron microscopy (TEM, SEM), it is possible that discrepancies in interpretation of preserved tissue by specialists in this field have been the reason why no text carries a clear description of the tissue organization of these sponges. Despite several excellent articles describing fixed tissues from some eight species of hexactinellids (e.g. Mackie & Singla 1983; Reiswig & Mehl 1991; Boury-Esnault & Vacelet 1994), a clear, three-dimensional picture of their tissues is still lacking. In other sponge groups the flagellated chambers and This content downloaded from 157.55.39.153 on Mon, 19 Sep 2016 04:45:05 UTC All use subject to http://about.jstor.org/terms

Ontogenies of Phototactic Behavior and Metamorphic Competence in Larvae of Three Species of Bugula (Bryozoa)

Abstract: The free swimming larvae of many marine invertebrates actively respond to light. Light cues can be used to regulate position in the water column and to facilitate encountering sites suitable for metamorphosis. We examined the ontogeny of larval phototaxis and the ontogeny of metamorphic competency in larvae from three congeneric species of bryozoans. Larvae of Bugula neritina are positively phototactic on emergence from the brood chamber, whereas larvae of B. simplex and B. stolonifera appear initially photoneutral when populations of larvae are examined. Larvae of all three species become photonegative with time. Temporally coincident with this change to negative phototaxis is an increase in the competency of larvae to initiate metamorphosis. This observation suggests that these events are either physiologically linked or co-occurring, but independent developmental processes. We tested these hypotheses by artificially changing the sign of phototaxis from positive to negative using 10-5 M bathapplied 5-hydroxytryptamine (5HT) in larvae of B. neritina that were swimming for 1 h. Larvae that were photopositive and 1-h-old did not metamorphose at levels significantly different from larvae that were 1-h-old and treated with 5HT (i.e., young, photonegative larvae). Additionally, photopositive larvae which were swimming for 4 h initiated metamorphosis at rates nearly identical to photonegative larvae of the same age. Our data document that in larvae of B. neritina the changes in sign of phototaxis and levels of metamorphic competency are independent developmental events that occur in temporal coincidence. The concurrent timing of these two pathways may have been synchronized through selective processes resulting in a tight coupling between arrival at potentially suitable sites for metamorphosis and ability to respond to metamorphic cues. Additional key words: bryozoans, phototaxis Many benthic marine invertebrates have as part of their life cycle a planktonic larval stage. Successful completion of the life cycle requires return to a suitable benthic habitat. Larvae utilize a variety of physical and biological cues to facilitate location of sites favorable for subsequent adult life (e.g., Pawlik 1992; Young 1995). Light is one such physical signal. Light is a ubiquitous vector in relatively shallow waters that exhibits both spatial and temporal variation. The temporal variation can be highly predictable (sunrise and sunset) or highly variable (changes in cloud cover). Responses to light are one potentially useful means of regulating vertical orientation and position in the water column (e.g., Thorson 1964; Clarke 1970; Cronin & Forward 1979; Sulkin 1984; Forward 1988; Barile et al. 1994; Young 1995). Additionally, in the case of meroplanktonic larvae, phototactic responses can condu tribute to delivery of larvae to benthic sites suitable for metamorphosis (e.g., McDougall 1943; Ryland 1960; Crisp 1974; Young & Chia 1982; Olson 1985; Dirnberger 1993). Thorson's (1964) seminal work was the first detailed comparative study of the ontogeny of phototaxis across several phyla of benthic marine invertebrates. He observed that 82% of 141 species from 11 phyla had early-stage larvae that were initially photopositive. Of these, 76% became photonegative before the conclusion of larval life. Thorson suggested that earlystage positive phototaxis increases the chances that larvae will be transported into the water column and hence away from conspecific adults and benthic predators. In the water column, larvae can be passively dispersed by currents, and feeding larvae can have access to more abundant populations of phytoplankton. Larvae that change to photonegative with time will then likely move to benthic sites, which is essential for completion of the life cycle. ies tactic ior t rphic tence e i s la ( a)

Pigmentation and Acid Storage in the Tunic: Protective Functions of the Tunic Cells in the Tropical Ascidian Phallusia nigra

Abstract: The tunic morphology of the tropical ascidian Phallusia nigra was studied with regard to its protective functions by means of light and electron microscopy. The tunic has a smooth surface free of epibionts. Beneath the tunic surface, phagocytic cells containing pigment granules (pigmentary tunic phagocytes) are densely distributed and form the pigmentary layer. These pigments may protect the body from irradiation. When poorly pigmented specimens growing in a dim location are transplanted to a bright location, pigmentation is induced and the tunic becomes black. In this process, tunic phagocytes probably produce pigment granules and migrate to the pigmentary layer. The tunic under the pigmentary layer is packed with highly vacuolated cells (tunic bladder cells). Vital staining with LysoSensor, a fluorescent pH indicator, demonstrated that the vacuoles contain strong acid. Acidic fluid leaks out when the tunic is injured. Leaking acid could serve to disinfect the injured area, be an irritant to potential predators, and/or be involved in anti-fouling. Additional key words: phagocytosis, pigment granules, light, protection, anti-predation, antifouling An integumentary tissue has a variety of functions to protect the body from the environment. It can be the most important defense system in many sessile organisms. The only sessile chordates are ascidians, whose bodies are wholly covered with a leathery or gelatinous matrix called the tunic. Although the tunic is a kind of extracellular matrix situated outside the epidermis, free cells called tunic cells are distributed within it, and the presence of tunic cells makes it a mesenchyme-like tissue. Many types of tunic cells have been described in ascidians: they are involved in phagocytosis (De Leo et al. 1981; Hirose et al. 1994), conduction of impulses (Mackie & Singla 1987), tunic contraction (Hirose & Ishii 1995), bioluminescence (Chiba et al. 1998), photosynthetic symbiosis (Hirose et al. 1996), and allorecognition between colonies (Hirose et al. 1997). The ascidian tunic is, therefore, an attractive material for investigating the diversity of integumentary functions required for adaptation to the environment. In the tropics, sessile organisms can expand potena E-mail: euichi@ sci.u-ryukyu.ac.jp tial habitat through protective systems against strong sunlight (cf. Jokiel 1980; Bingham & Reyns 1999) as well as other environmental pressures, such as predation, infection, and competition for space. Phallusia (= Ascidia) nigra SAVIGNY 1816 is a large solitary ascidian inhabiting shallow tropical waters. Its tunic and mantle contain black pigment that may protect the body from sunlight. Moreover, P. nigra is known to concentrate vanadium and to have an extremely acidic tunic fluid, and these properties may be effective for anti-fouling as well as anti-predation (Stoecker 1978, 1980a,b). The tunic morphology and chemical analysis have been reported for this species with special reference to anti-fouling function (Stoecker 1978), but ultrastructural studies had not been carried out. I describe here the morphology of the tunic cells in P. nigra and examine their possible protective functions in adaptation to the tropical environment. There are two types of tunic cells: one type contains phagosomes and pigment granules, and the other type stores strongly acidic fluid in large vacuoles. These features may be concerned with protection against infection (or scavenging), light, and predation, respectively. This content downloaded from 157.55.39.248 on Sun, 12 Mar 2017 18:08:46 UTC All use subject to http://about.jstor.org/terms Pigmentation and acid storage in ascidian tunic

Paleochordodes protus n. g., n. sp. (Nematomorpha, Chordodidae), parasites of a fossil cockroach, with a critical examination of other fossil hairworms and helminths of extant cockroaches (Insecta: Blattaria)

Abstract: Two hairworm parasites of a cockroach in Dominican amber (15-45 mya) are described as Paleochordodes protus n.g., n.sp. (Nematomorpha, Chordodidae) and represent the first unequivocal record of fossil nematomorphs. Previous reports of putative nematomorphs are discussed and shown to be tentative at best. A critical examination of supposed mermithid nematode infections of cockroaches concludes that none of these can, at present, be confirmed. "Worms arise from the fat of roaches" Pliny the elder, Natural History, Chapter 6 (AD 77) Nematomorpha comprises a group of invertebrate parasites that are unique in many aspects (Poinar 1991 a,b). Literature dealing with fossil nematomorphs is rare and consists only of reports by Voigt (1938) and Xianguang & Bergstrom (1994), both of which are critically examined here and found to be tentative. A piece of amber from the Dominican Republic was discovered to contain two hairworms, one partly and one fully emerged from their cockroach host (Fig. 1). These rare specimens are described below in a new genus and species and a synopsis of the helminth fauna of extant cockroaches is presented.

Ultrastructure of the Body Cavity Lining in Embryos of Epiperipatus biolleyi (Onychophora, Peripatidae): A Comparison with Annelid Larvae

Abstract: In contrast to the Annelida, no trace of the developing coelomic cavities persists in the body cavity of postembryonal stages of Onychophora and Euarthropoda; instead, a primary body cavity or haemocoel is characteristic for these stages. Transitory coelomic cavities, which appear in certain developmental stages of arthropods, are generally regarded as recapitulation of paired and segmental coelomic cavities of the common ancestor of annelids and arthropods. If this is true, a specific organization of these cavities with respect to the annelid coelom is to be expected. In order to test this assumption, the transitory coelomic cavities of the onychophoran Epiperipatus biolleyi are investigated ultrastructurally and compared with data on the coelomogenesis of the polychaetous annelids Ophelia rathkei, Spirorbis spirorbis, and Pectinaria koreni. The somatic wall of the transitory coelom of Epiperipatus biolleyi is thicker than the visceral (splanchnic) wall; the lining cells are apically connected by adhaerens junctions and possess a rudimentary cilium, and are otherwise largely undifferentiated. The latter observation is clearly in contrast to comparably differentiated coelomic cavities in annelids. During further development the haemocoel is formed in Epiperipatus biolleyi. A survey of the older literature reveals that in the Onychophora the haemocoel is formed by fusing interstices within the subepidermal and peri-intestinal matrix, and not by a fusion of such interstices and the

Halicryptus higginsi n.sp. (Priapulida): A Giant New Species from Barrow, Alaska

Abstract: A new species of Priapulida was discovered from the Beaufort Sea near Barrow, Alaska. Following an unusually strong autumnal storm, 30 adult specimens of Halicryptus higginsi n.sp. were collected, many alive, from intertidal beaches Additional specimens were found in museum collections misidentified as Halicryptus spinulosus von Siebold 1849. The new species represents the 11th priapulid species described since 1968 and increases the number of described extant species of Priapulida to 18. While all other recently described priapulids have been meiofaunal, Halicryptus higginsi n.sp. is macrofaunal and the largest extant priapulid species, with one specimen being 39 cm long in a contracted condition. Additional key words: arctic, benthos In recent years the concept of the marine phylum Priapulida has changed radically, from that of a macrobenthic, cold-water taxon, to one that is more speciose in the tropical meiofauna. Before 1968, only 7 extant species were known and additional species had been described from the fossil record; all extant species were macrobenthic and found only in cold-water habitats (Higgins et al. 1993). In 1968 the first meiobenthic priapulid, Tubiluchus corallicola VAN DER LAND 1968, was described from a tropical, shallowwater habitat (van der Land 1968). Since 1968 an additional 9 priapulid species, all meiobenthic and mostly inhabitants of tropical, shallow-water sediments, have been described. Tubiluchus arcticus ADRIANOV, MALAKOV, CHESUNOV, & TSETLIN 1989 is the only cold-water species that has been described in recent years, and it is a meiofaunal priapulid that inhabits shallow-water sediments in the arctic (Adrianov et al. 1989). During a survey of museum specimens of priapulids, one of us (VS) discovered that specimens in the United States National Museum of Natural History, Smithsonian Institution, misidentified as Halicryptus spinulosus von Siebold from Barrow, Alaska, had been used for scanning electron microscopy (Merriman 1981) and instead represented a new and undescribed congener. Subsequent to the museum discovery we found larvae and juveniles of the undescribed species a Author for correspondence. E-mail: fftcs@uaf.edu in the shallow subtidal during two expeditions to Barrow, Alaska in 1991 and 1992, but did not find adults of the new species. Following an unusually strong autumnal storm in 1993, 30 adult specimens of Halicryptus higginsi n.sp. were collected, many alive, from intertidal beaches near Barrow. In this paper we describe the new species and compare it with H. spinulosus. The larvae will be described in a separate pub-

Deepest Atlantic Molluscs: Hadal Limpets (Mollusca, Gastropoda, Cocculiniformia) from the Northern Boundary of the Caribbean Plate

Abstract: Descriptions and scanning electron micrographs of the shell, protoconch, radula, and gross external anatomy are provided for four species of cocculiniform limpets collected from hadal depths of the Cayman and Puerto Rico trenches. Of these, one represents a new genus and species, Macleaniella moskalevi (Cocculinidae), endemic to the Puerto Rico Trench. Another, Amphiplica plutonica (Pseudococculinidae), represents a new species from the Cayman Trench of a globally distributed abyssal genus. The original descriptions of the remaining two species, Fedikovella caymanensis (Cocculinidae) and Caymanabyssia spina (Pseudococculinidae), both from the Cayman Trench and type species of their respective genera, contained only poorly reproduced illustrations of portions of the shells, and line drawings of isolated rows of radular teeth. These four species, each apparently endemic to their respective trenches, represent the deepest records of molluscs collected in the Atlantic Ocean. Additional key words: Deep-sea, Cocculinidae, Pseudococculinidae Cocculiniform limpets have remained one of the Harasewych (1995) provided morphological data on most taxonomically perplexing and enigmatic groups previously described and new bathyal and abyssal speof gastropods since their first discovery at bathyal cies, but were unable to examine representatives of the depths along the western Atlantic over a century ago two known hadal taxa. Subsequently, we were able to (Dall 1882). Since then, the accelerating pace of deepstudy samples from 24 hadal bottom trawlings on the sea sampling, especially through the use of research floors of the Cayman and Puerto Rico Trenches that submersibles, has greatly increased the number of were collected by the University of Miami research known cocculiniform taxa. This, in turn, has prompted vessels John Elliott Pillsbuiy and James M. Gilliss growing interest in all aspects of the systematics and during the late 1960s and 1970s. Collected as part of biology of these animals (e.g., Moskalev 1976; Hickthe National Geographic SocietyPJational Science man 1983; Marshall 1986; Haszprunar 1987, 1988a,b; Foundation-funded project \"Investigations of the BiMcLean 1987, 1988, 1991, 1992; Dantart & Luque ology and Distribution of the Tropical Deep-sea Fau1994; McLean & Harasewych 1995; Haszprunar & na\" (Table 1; Voss 1973, 1976), the samples were deMcLean 1996). Today, well over a hundred species are posited in the Marine Invertebrate Museum, Rosenstiel divided among two superfamilies and ten families, School of Marine and Atmospheric Science, Univerwith each family restricted to one or a very few biosity of Miami (UMML). Similar explorations in the genic substrates that sink into the deep sea (see Hasame two trenches were carried out during Cruise 14 szprunar 1988a). Of these, the best known and most of the Russian vessel Akademik Kurchatov in 1973 diverse are the Cocculinidae (superfamily Cocculino(Pasternak et al. 1975; Wolff 1979). Aside from brief idea) and the Pseudococculinidae (superfamily Lepetelloidea), by virtue of their association with sunken references in articles on deep-sea ecology (e.g., wood and plant material, which are more easily samGeorge & Higgins 1979; Wolff 1979, 1980), the bulk pled than other substrates. of the UMML samples, which include hadal represenIn a recent review of the cocculinid and pseudococtatives of several invertebrate groups, remained essenculinid limpets of the western Atlantic, McLean & tially untouched. Four species of cocculiniform limpets were present in 10 of these 24 trawlings. The species referable to the family Cocculinidae are Fe\" Author for correspondence. E-mail: leal@water.net dikovella caymanensis MOSKALEV, 1976 and Maclean-117 Hadal cocculiniform limpets Table 1. Hadal stations occupied by the University of Miami R N s John Elliott Pillsbziry (P) and James M. Gilliss (GS) along the northern boundary of the Caribbean Plate. Locations are Puerto Rico Trench (PRT), north of Puerto Rico Trench (NPRT) and Cayman Trench (CT). Depths are corrected. Locations represented by stations P-1406 and GS-121 are also respectively known as Milwaukee Deep and Oriente Deep. Station P-1376 is not hadal, but provided specimens of Macleaniella moskalevi. Gear used was either 10' Blake trawl (BT) or 41' otter trawl (OT). Signs (+) indicate stations yielding benthic molluscs, and (++) stations yielding cocculiniform limpets. Cruise Station Location Coordinates Depth (m)

Babysitting brittle stars: heterospecific symbiosis between ophiuroids (Echinodermata)

Abstract: Juveniles of Ophiomastix annulosa live on adults of Ophiocoma scolopendrina, a confamilial broadcast-spawner, in the intertidal zone of Sesoko Is., Okinawa, Japan. This is the first reported instance of a symbiosis between heterospecific ophiuroids. In the sense that the juveniles are brooded by another species, they may be regarded as "brood parasites." There is no evidence that the symbionts physically damage the host, although they may steal its food. 0. annulosa of less than 4.5 mm disk diameter primarily occupy the bursae, and also cling to the arms or disk of their hosts. The association is seen year round, usually with 4.2% or fewer of 0. scolopendrina being involved. There is generally one juvenile, rarely two, on a host. The body size of the host may limit the maximum size of its symbionts. The ontogenetic transition of 0. annulosa from a symbiotic to a free-living life style involves marked changes in integument, arm spine morphology, tube foot structure, disk armament, and pigmentation pattern. Adults of Ophiocoma scolopendrina harbor conspecific juveniles as well, a relatively rare behavior among ophiuroids, but only juveniles of 0. annulosa occur in its bursae, the respiratory structures that in brooding and viviparous ophiuroids also serve as brood chambers. An association between adult and juvenile individuals is documented for a Panamanian population of Ophiocoma aethiops, and evidence is presented of a heterospecific symbiosis between juveniles of Ophiomastix janualis and adults of Ophiomastix flaccida in the Philippines. Additional key words: brood parasitism, juvenile adaptations, Amphipholis squamata, Ophiomastix spp., Ophiocoma spp. Meager information exists on the postlarval period of the ophiuroid life cycle, when, as in other marine benthic invertebrates, mortality commonly exceeds 90% (Hendler 1991; Gosselin & Quian 1997). Several studies of juvenile ophiuroids suggest that characteristics of their microhabitat, and in some cases the presence of adult ophiuroids, can be crucial to their survival (Hendler 1991). Some juvenile ophiuroids are abundant in algal "refuge" substrates, where the adults and juveniles of brooding species co-occur (Hendler & Littman 1986). Adult epizoic ophiuroids and their young often occupy the same sponge and coral hosts (Clark 1976; Hendler 1984). In addition, the juveniles of some broadcast-spawning species cling to conspecific adults and thereby become established in a suitable milieu (Hendler 1991). We describe an unexpected association between jua Author for correspondence. E-mail: hendler@nhm.org b Present address: Lake Biwa Museum, 1091 Oroshimo-cho, Kusatsu-shi, Shiga-ken 525-0001, Japan. veniles of Ophiomastix annulosa (LAMARCK 1816) and adults of Ophiocoma scolopendrina (LAMARCK 1816). 0. scolopendrina is abundant and widespread throughout most of the Indo-Pacific region. Its symbiont, 0. annulosa, has a narrower range, from the Maldives (Indian Ocean) to the South Pacific region (Devaney 1970; Clark & Rowe 1971). 0. scolopendrina differs from other Ophiocoma and Ophiomastix species in its restriction to sheltered intertidal habitats (Sloan et al. 1979; Liao & Clark 1995). We report that the symbiotic stage of 0. annulosa is also intertidal, although the adults and those of its congeners typically occur subtidally (Devaney 1978; Liao & Clark 1995). This is the first account of heterospecific symbiosis between ophiuroids. The relationship between 0. annulosa and 0. scolopendrina is more intimate than the association between the adults and conspecific juveniles of 0. scolopendrina, and it may be regarded as a form of brood parasitism. We illustrate the ontogeny of the juveniles, which are unlike the adults in appearance, and discuss several morphological features This content downloaded from 157.55.39.45 on Fri, 02 Sep 2016 05:31:33 UTC All use subject to http://about.jstor.org/terms Brittle star symbiosis that may be related to their symbiosis. In addition to the Ophiocoma-Ophiomastix association, we document the presence of juveniles of Ophiomastixjanualis LYMAN 1871 on adults of Ophiomastixflaccida LYMAN 1864 and report the association of juvenile conspecifics with the adults of Ophiocoma aethiops LUTKEN

Sweeper polyps of the coral Goniopora tenuidens (Scleractinia : Poritidae)

Abstract: The massive coral Goniopora tenuidens can develop elongated sweeper polyps. These are thought to be involved in aggressive interactions with neighbouring benthic organ isms, like the sweeper tentacles of other corals. The cnidoms of sweeper polyps and ordinary polyps of G. tenuidens from the Great Barrier Reef were compared. Sweeper polyps had significantly greater densities of elongate holotrichous isorhizas (34577 +/- 3839/mg; mean +/- SD, n = 6) than ordinary polyps (936 +/- 371/mg; p < 0.05), while ordinary polyps had significantly greater densities of spirocysts (75994 +/- 15992/mg) than sweeper polyps (19469 +/- 7808/mg; p < 0.05). This suggests that sweeper polyps of G. tenuidens, like the sweeper tentacles of other corals, are modified for aggression, and that they probably act through nematocyst discharge. However, the scattered distribution of sweeper polyps observed on colonies of G. tenuidens in the field suggests that sweeper polyps may have other functions.

Filter Feeding in Lancelets (Amphioxus), Branchiostoma lanceolatum

Abstract: Little work has been done to describe lancelets as filter feeders, and hitherto no data have existed on either filtration rates or the efficiency with which food particles are retained in the filters. The aim of the present work was to describe and characterize Branchiostoma lanceolatum as a true filter-feeding species. The clearance-rate capacity in B. lanceolatum was measured at naturally low algal concentrations, and used as a reference state representing the normal filtration rate in nat ure and for comparisons with video recordings of the feeding currents. Simultaneous measurements of clearance different-sized particles, prepared as a mixture of flagellates (Isochrysis, Rhinomonas, and Tetraselmis) showed that particles ?4 pLm are retained by the mucus filter with 100% efficiency. The wavelength and speed of the metachronal wave of the lateral cilia, which generates the power of the branchial pump, were measured to estimate the ciliary beat frequency. Furthermore, transport rate, production, exposure time, and flow speed of water through the mucus filter was estimated. The loss of body dry-weight during a long term starvation period was used as an indirect measure of the respiration rate. The results are used to characterize filter feeding in B. lanceolatum in a way that may be compared to a number of other macro-invertebrate filter feeders that have been studied during recent years. Additional key words: water pumping, clearance, video recordings, retention efficiency, pump design Filter feeding lancelets may superficially resemble small primitive fishes with their dorsal fin and gill slits. They constitute a small phylum (or subphylum) (Cephalochordata), comprising only -25 marine species (Poss & Boschung 1994; Nielsen 1995). One of the most common of these transitional species, which share characteristics with both vertebrates and invertebrates, is Branchiostoma lanceolatum, generally known as amphioxus. This animal has been the focus of evolutionary studies for more than a century, and the bibliography of lancelets includes some 2700 references (Gans 1996). The cephalochordates and vertebrates are monophyletic taxa of the Chordata, but the two separate phyla should be regarded as sister groups (Nielsen 1995). Since the early contributions of Orton (1913) and Weel (1937), a number of papers have described lancelets as filter feeders (Barrington 1958; Olsson 1963; Welsch 1975; Baskin & Detmers 1976; Rahr 1982) but no data exists on either filtration rates or the efficiency with which food particles are retained in the filters (for a review of the early literature, see J0rgensen 1966). a Corresponding author. E-mail: hur@biology.ou.dk The aim of the present work is to further describe and characterize members of B. lanceolatum as true filter feeders which may be compared to the many macroinvertebrate filter feeders. A great variation in the design of biological filter-pumps has evolved to solve the same basic problem encountered by these animals, namely the necessity of extracting sufficiently large amounts of microscopic food particles from highly dilute suspensions (Riisgfard & Larsen 1995). But this has also led to a great number of similarities between widely different species of suspension feeders. The present work aims to accentuate such recognition. As obligate filter feeder, individuals of B. lanceolatum obtain food by straining nutritive particles, mainly phytoplankton, from the surrounding water. When feeding, the lancelet lies completely or partly buried in the bottom gravel with the ventral side turned upward and with the mouth opening free of the bottom (Fig. 1A,B). A feeding current enters at the anterior end of the animal between the buccal tentacles, which act as a coarse sieve, and flows through the vestibule (buccal cavity). The water then flows successively through the mouth, surrounded by a circular velum and attached velar tentacles, the branchial basket (pharThis content downloaded from 157.55.39.137 on Fri, 27 May 2016 05:34:38 UTC All use subject to http://about.jstor.org/terms

Programmed Cell Death during Longitudinal Fission in a Sea Anemone

Abstract: Little is known about the cellular events regulating fission in metazoans. In sea anemones, longitudinal fission begins with stretching of the body column and culminates in ripping apart of the animal. Previously, we found that mechanical stretching of the animal plays a regulatory role in early fission events. In this study we use histology, TUNEL cytochemistry, and TEM to analyze the possible spatio-temporal relationship between stretching of tissues during fission and programmed cell death (apoptosis) within stretched tissues. We report that enhanced apoptosis occurs in specific tissue regions apparently most affected by stretching during fission. In stretched animals we find a significant induction of apoptosis at the junctions of body wall and particular mesenteries that begins in the axis parallel to stretch and then progresses to the axis perpendicular to stretch as fission progresses. Based on these results, we propose a model whereby stretching induces apoptosis in populations of cells, allowing tissue to thin and thus facilitating successful fission. Additional key words: Anthozoa, Cnidaria, Haliplanella, apoptosis, asexual reproduction,

Adaptive Function of Gastropod Larval Shell Features

Abstract: Similarities in features that occur commonly on the shells of many planktotrophic gastropod larvae may be interpreted primarily in terms of (a) shared ancestry, (b) shared adaptive solutions to problems of planktonic existence, or (c) shared artifacts of a basic pattern of early biomineralization. The Paradigm Method of functional inference, developed by paleontologists to analyze structures of fossil organisms, is used to analyze recurring features of unknown function on gastropod larval shells. Four recurring features are shown to fit paradigmatic adaptive solutions to preventing or retarding mechanical breakage at the growing apertural margin of the larval shell. These features are apertural beaks, velar notches, peripheral angulations, and continuous spiral elements of shell sculpture. A fifth common feature on larval shells, discrete crystalline microprotuberances, is inferred to be primarily a non-functional artifact of remote biomineralization and self-organized growth, although secondary functional advantage cannot be discounted. Protective functional interpretations are supported by patterns of shell breakage and repair on larvae from the plankton and by patterns of shell failure in trials subjecting larval shells to artificial mechanical attacks. The ability to withstand mechanical onslaughts of zooplankton predators and to repair damage to the larval shell may be a highly significant means of reducing natural larval mortality. Additional key words: Paradigm Method, analysis of form, biomineralization, predation,

\emph{Smithsoninema inaequale} n.g. and n.sp ({Nematoda, Leptolaimidae}) inhabiting the test of a foraminiferan

Abstract: Smithsoninema inaequale n.g., n.sp. (Nematoda: Leptolaimidae: Camacolaiminae) is described from one male and one female specimen taken from within an arenaceous foraminiferan, Vanhoeffenella aff. gaussi RHUMBLER 1905 (Foraminifera: Astrorhizidae) collected off the coast of Luanda, Angola. The male has a cuticularized, stylet-like thickening of the dorsal buccal wall, and the posterior end of the esophagus is glandular and elongated as in members of Anguinoides and Onchium (Camacolaiminae). Smithsoninema n.g. is characterized by its strong sexual dimorphism. Males are long, slender, and structurally similar to other males of Camacolaiminae. Females are short and stout, with a voluminous, trophosome-like midgut devoid of an internal lumen. The rectum and anus are vestigial. Through the transparent walls of the test it was possible to see the male and female nematodes enclosed within the theca of the foraminiferan. The nematodes were probably parasites, or endo-predators, rather than inquilines. Camacolaiminae is redefined and Nemella COBB 1920 is synonymized with lonema COBB 1920. Dagda SOUTHERN 1914 and Diodontolaimus SOUTHERN 1914 are transferred to Camacolaiminae, although Dagda asymmetrica GERLACH 1953, D. phinneyi MURPHY 1964, and Diodontolaimus tenuispiculum GERLACH 1955 are transferred from their respective genera to Procamacolaimus. Eontolaimus FURSTENBERG & VINCX 1988 is synonymized with Listia BLOME 1982 (Leptolaiminae). A key to the genera of Camacolaiminae is provided. Additional key words: Camacolaiminae, onchiostyle Several collections of deep-sea nematodes taken on cruises of the Woods Hole Oceanographic Institute's research vessels have been deposited at the National Museum of Natural History, Smithsonian Institution. Included among these collections are nematodes taken during cruise 42 of Atlantis II, from Walvis Bay, Namibia to Luanda, Angola. An examination of this material during a visit to the National Museum of Natural History by the second author revealed a single foraminiferan test containing one male and one female of a new genus and new species, described herein.

Distinguishing Early Juveniles of Eastern Pacific Mussels (Mytilus spp.) Using Morphology and Genomic DNA

Abstract: A lack of efficient criteria by which to discriminate among Mytilus spp. in early juveniles (-0.5-5.0 mm shell length) has hindered studies of recruitment, despite the conspicuous ecological roles played by the adults on rocky shores o n North America. We present morphological criteria, visible under a dissecting microscope, that distinguish sea mussels, Mytilus californianus, from sympatric bay mussels, M. trossulus and M. galloprovincialis. The shells of early juvenile sea mussels from southern British Columbia and from southern California have (1) a more posterior dorsal apex and (2) a posterior adductor muscle (PA) attachment larger than do shells of bay mussels from similar locations (M. trossulus and M. galloprovincialis, respectively). Accuracy of these criteria was confirmed by comparing PCR products of genomic DNA of juveniles with those of adults. Additional criteria useful in identification include: (3) the proximity of the PA to the dorsal apex line, (4) the dorsal angle of the dissoconch, and (5) lateral hinge-tooth demarcations. Morphological criteria identified by this study constitute a fast, economical means of distinguishing early juveniles of those mussels and may benefit ecological research on Mytilus. Additional key words: dissoconch, shell morphology, recruitment Since the early 1980s, ecologists have increasingly supported the proposition that recruitment is an essential feature of the dynamics of benthic communities (Keough 1984; Underwood & Denley 1984; Gaines & Roughgarden 1985; Gaines et al. 1985). Variation in densities of juvenile invertebrates produces dominant age classes (Dayton 1971; Keough 1984), modulates the intensity of species interactions (Menge et al. 1994; Robles 1997), and influences movements and standing stocks of predators (Menge 1992; Robles et al. 1995). Unfortunately, detailed study of recruitment may be hindered by the lack of simple criteria by which to identify animals at the early post-metamorphic or juvenile stages. Such a problem is encountered studying mussels of the genus Mytilus. The adults of three species, the sea mussel, Mytilus californianus CONRAD 1837, and the bay mussels (also called "blue" mussels), M. trossulus GOULD 1850 and M. galloprovincialis LAMARK 1819, play varied and important ecological roles along the a Author for correspondence. Present address: Bamfield Marine Station, Bamfield, B.C., VOR iBO, Canada. E-mail: amartel@bms.bc.ca western coast of North America (Paine 1966, 1974, 1976; Suchanek 1981, 1985). However, the distinguishing morphological features of the adults are not differentiated in early juvenile stages. Indeed, while presenting molecular criteria for juvenile identification, Heath et al. (1996) repeat the view (Suchanek 1978) that the juveniles of bay mussels <10 mm long cannot be physically distinguished from those of sea

Parasites of Larval Black Flies (Diptera: Simuliidae) and Environmental Factors Associated with Their Distributions

Abstract: A total of 26,673 midto late-instar larval black flies were collected from 115 stream sites in South Carolina. Host larvae with patent, externally visible infections of nematodes, fungus, or microsporidia were identified to species either morphologically or cytologically. Six parasite taxa were identified: nematodes of the family Mermithidae; the fungus Coelomycidium simulii; and the microsporidia Janacekia debaisieuxi, Polydispyrenia simulii, Amblyospora bracteatalvarians, and Caudospora simulii. Of 43 species of black flies, 25 harbored at least one of these parasite taxa. Records from museum specimens increased the number of infected host species in South Carolina to 34 species. Prevalence of patent infections for each parasite taxon was low, ranging from 0.1 to 6.5% of all larvae. The distribution of nematodes among stream sites was nonrandom and correlated with season, host abundance, and stream-site conditions. The association of nematode distributions with stream parameters might reflect environmental influences on the nematodes during their free-living period and on host susceptibility to infection. Additional keywords: Mermithidae, Coelomycidium simulii, microsporidia Larval black flies (Diptera: Simuliidae) are a major component of the macroinvertebrate fauna in streams and rivers (Adler & McCreadie 1997). They anchor themselves to a silken pad spun on solid substrates (e.g., rocks, trailing vegetation) and filter food from the water column or graze it from the substrate. Larval and pupal development is temperature dependent and requires from one week to half a year (Crosskey 1990). After emergence from the pupa, females of most species mate, acquire sugar for energy and a vertebrate bloodmeal for egg maturation, and find a suitable stream for oviposition. Larvae are hosts of a variety of parasites, especially mermithid nematodes (Mermithidae), microsporidian protozoans (Microspora), and the chytrid fungus Coelomycidium simulii DEBAISIEUX. Preparasitic mermithid nematodes crawl about on the stream substrate and use a protrusible stylet to penetrate the host body (Molloy 1981). Infected host larvae are recognized by the presence of one or more coiled worms visible through the abdominal cuticle. Mermithids either exit and kill the host while it is still in the larval stage or pass into the adult stage, exiting shortly thereafter (Crosskey 1990). Postparasitic worms molt to adults, mate, and deposit eggs in the streambed (Poinar 1981). a Author for correspondence. E-mail: jmccrd@clemson.edu Microsporidian infections cause pathological changes, primarily to the host fat body (Weiser & Undeen 1981). Patent infections are recognized by the presence of large, irregular cysts that distend the host abdomen. Life cycles of microsporidia that attack black flies are not well understood, but larval hosts become infected both by transovarial transmission and by ingesting the free-living spores (Crosskey 1990). Spores might require some form of environmental conditioning before they become infective, and alternate hosts might be required to complete the life cycle (Jamnback 1973; Lacey & Undeen 1987). Larvae with patent infections of the fungus C. simulii are packed with minute, spherical thalli throughout the body cavity. Thalli give rise to spores that are released into the water column after the death of the host (Tarrant 1984). An intermediate host might be required to complete the life cycle (Lacey & Undeen 1987). In the present study, we survey selected parasites of larval black flies in South Carolina. Given that nematodes have free-living stages, we ask whether these parasites occupy a restricted range of stream conditions within the broader range of conditions under which their hosts are found. In other words, can occurrence (presence/absence) of nematodes in a particular host species, at a particular location, be predicted This content downloaded from 157.55.39.84 on Wed, 15 Jun 2016 06:30:20 UTC All use subject to http://about.jstor.org/terms Parasites of larval black flies by local stream conditions? Our objective is to examine factors that can explain parasite distributions, with the tenet that patterns of distribution can provide insight into the basic ecology of these organisms. Other parasites (microsporidia, C. simulii) were collected in numbers too low to address these questions.

Fine structure and evolutionary significance of sagittocysts of Convolutriloba longifissura (Acoela, Platyhelminthes).

Abstract: Sagittocysts are extrusomes found only in acoel turbellarians. They are needlelike secretory products, on the order of 18-50 pxm long and 1-5 pxm wide, and consist of a fibrous cortex, a central filament, and an intermediate lucent layer. We discovered sagittocysts in Convolutriloba longifissura, for which they had not been known before, by using confocal microscopy and a phalloidin-conjugated fluorescent stain that strongly labeled a mantle of muscle around the distal neck of the sagittocyst-secreting cell, the sagittocyte. The muscle mantle apparently plays a role in ejecting the sagittocyst. Positions of the sagittocysts revealed by confocal microscopy suggests their role in defense as well as in prey capture. By electron microscopy, the differentiation of sagittocysts was evident in the proximal part of the sagittocyte. The muscle mantle on the neck of the sagittocyte is conical in shape; its ribbon-like myocyte enwraps the neck in a tight spiral. Extrusion of sagittocysts could be induced by stimulation with electrical pulses, light pulses, or weak hydrochloric acid; only the whole sagittocyst was ejected, not its central filament. The presence of sagittocysts in species of Convolutriloba is sufficient to reassign the genus to the family Sagittiferidae. We establish the new subfamily Convolutrilobinae, with 3 species in the genus Convolutriloba, and new subfamilies for the remaining sagittiferids as well. As extrusomes, sagittocysts are comparable to nematocysts, colloblasts, rhabdites, and other extrusomes common especially to lower eumetazoans, and the origin of all such extrusomes may correlate with the origin of the eumetazoan

Glial Cells of the Central Nervous System in the Crab Ucides cordatus

Abstract: Glial cells and their processes were characterized in the fasciculated zone and in the protocerebral tract of the crab Ucides cordatus by light and electron microscopy. Thiery and PAS procedures indicate the presence of carbohydrates, particularly glycogen in cells. Immunohistochemistry was used to observe tubulin distribution in the glial cells. Our results demonstrate at least two types of glial cells in the fasciculated zone and in the protocerebral tract, separable by their location and electron density. Judging by their position, electron-lucent cells may correspond to periaxonal cells and electron-dense ones may correspond to perineurial cells. The electron-dense processes have previously been interpreted as extracellular matrix, but since they feature an enveloping membrane and contain glycogen and mitochondria (intact and with varying degrees of disruption) we consider them to be part of one type of glial cells. Additional key words: Crustacea, Malacostraca, Brachyura, glia Invertebrate glial cells have traditionally been classified according to certain relatively general morphological or functional criteria and also by their anatomical position (Haimori & Horridge 1966; Radojcic & Pentreath 1979). Because invertebrates do not have true compact myelin, no glial class equivalent to vertebrate oligodendrocytes can be identified. In addition, immunocytochemical markers for vertebrate astrocytes in most cases fail to label invertebrate glia. (One of the exceptions is glutamine synthetase, a specific marker for astrocytes in vertebrates, which has also successfully labeled non-neuronal cells in lobster olfactory regions: Linser et al. 1997.) The glial cell types of arthropods share many morphological characteristics. First, in many species, the chromatin of glial cell nuclei is clumped in the periphery. This feature is rarely seen in neuronal nuclei. Second, mitochondria, endoplasmic reticulum, and Golgi structures are generally common. Third, between neighboring glial cells and also between glial cells and neurons, diverse membrane specializations are found (see Pentreath 1987). In the nervous system of some species of higher crustaceans, i.e., Decapoda, there is good anatomical and physiological evidence for an insulating sheath with nodes and rapid saltatory conduction (Heuser & Doggenweiler 1966). Where present, the sheath coma Author for correspondence. E-mail: sallodi@chagas.biof.ufrj.br prises numerous laminae, each containing cytoplasm and forming a seam. The nodes are loosely wrapped by a characteristic type of glial cell called the nodal cell. The axons can be wrapped either simply by a single glial process or in groups, by complex, tightly packed multiple layers, resembling vertebrate myelin (Bullock et al. 1977). Important differences from vertebrate myelin are: (a) the crustacean laminae do not connect to form a spiral; (b) the nuclei of the sheath cells lie on the inside of the sheath; (c) desmosomes join adjacent laminae; (d) sometimes an adjacent extracellular matrix may occur between neighboring glial cell processes (Heuser & Doggenweiler 1966). The participation of neuroglia in the blood-brain barrier is an important function of glial cells in both vertebrates and invertebrates (Abbott et al. 1986; Abbott 1995). Unlike the situation in higher vertebrates, in which the blood-brain barrier is formed by tight junctions of endothelial cells lining capillaries of the central nervous system (see Peters et al. 1991), the barrier in crustaceans is formed by glial perineurium surrounding the ganglia. The sealing function of the barrier is thought to be due to the junctions between the perineurial cells and the extracellular matrix, which may be expanded and contain collagen-like fibrils (Abbott 1972; see Pentreath 1987). The most widely accepted function of glial cells is that they interact metabolically with neurons, providing nutrients, exchanging metabolites, and removing catabolites. Such a role is supported by glial stores of glycogen This content downloaded from 157.55.39.231 on Wed, 05 Oct 2016 04:16:00 UTC All use subject to http://about.jstor.org/terms Allodi, Silva, & Taffarel and large numbers of vesicular and granular inclusions in the cytoplasm (Pentreath 1987; Tsacopoulos & Magistretti 1996). Our goal was to characterize the organization of glial cells and their processes by light microscopy (LM) and transmission electron microscopy (TEM) and to correlate those findings with the distribution of extracellular matrix in the fasciculated zone (FZ) and in the protocerebral tract (PT) of the crab Ucides cordatus. The FZ comprises retinular cell axons that run from the retina to the lamina ganglionaris (Bell & Lightner 1988) and the PT is the tract that links the terminal medulla and the hemiellipsoid body to the anterior medial protocerebral neuropil and other areas of the brain (Sandeman et al. 1992). These neuropils were chosen because they are primarily composed of axons and glial cells.

Flat-Tipped Sensillum in Baetidae (Ephemeroptera): A Microcharacter for Taxonomic and Phylogenetic Considerations

Abstract: In some mayfly nymphs, family Baetidae, flat-tipped sensilla occur on antennal segments and on other body parts (tergal borders and cerci). These fine, flat-tipped setae have an apical pore, a basal tubular body, and a branched dendrite along the shaft, an organization consistent with a chemo-mechanosensory function. Phylogenetic considerations are presented on the basis of a survey on the presence or absence of flat-tipped sensilla in some members of the Baetidae. The results of this investigation, and those of other authors, support the hypothesis that this sensillum is an apomorphic character. Additional key words: mayflies, sensilla, phylogeny, taxonomy Insect sensilla have mainly been studied from two different perspectives: first, from the morphological and electrophysiological point of view, to examine their specific function (Crnjar & Prokopy 1982; Kapoor & Zachariah 1983; Keil & Steinbrecht 1984; Mclver 1985; Zacharuk 1985; Solinas et al. 1987; Stadler et al. 1994; Isidoro et al. 1996), and second, for the taxonomic relevance assigned to these structures, including setae, scales, and spines. In the Ephemeroptera, cuticular armatures represent a useful set of characteristics that taxonomists have considered in comparing different species. Scanning electron microscopy (SEM) has allowed new microcharacters to be highlighted. In this regard, Morihara & McCafferty (1979) used SEM in the taxonomic revision of the North American genus Baetis to examine and analyze the setae, scales, and spines of the nymphs. Among the three main types of setae studied by these authors, hairlike or bristlelike fine setae that were distinctly clubbed apically were considered to be phylogenetically important traits. This kind of fine sea Author for correspondence. E-mail: gaino@unipg.it tae has been described in the antennae of Baetis rhodani (Baetidae) and of Rhithrogena loyolaea and Epeorus sylvicola (Heptageniidae) (Gaino & Rebora 1996). The location of this setal type and the presence of an apical pore have supported the hypothesis that it could be a sensillum, defined as a flat-tipped sensillum (Gaino & Rebora 1996). Ultrastructural serial sections showed the presence of both a tubular body and the dendrites extending along the shaft, further supporting the hypothesis of a chemo-mechanosensory function (Gaino & Rebora 1998). Considering the possible systematic relevance of this sensillum to systematics, we have documented its presence (or absence) and its distribution in species representing almost all the European genera of the family Baetidae: Acentrella, Alainites, Baetis, Centroptilum, Cloeon, Labiobaetis, Nigrobaetis, Procloeon, and Pseudocentroptilum. For comparison, we examined the antennae of the New Zealand mayfly Siphlaenigma janae, belonging to Siphlaenigmatidae, generally regarded as the sister group of Baetidae. In this paper we report that flat-tipped sensilla are a Figs. 1-4. Flat-tipped sensilla in Baetis rhodani. Fig. 1. The sensillum connects two consecutive antennal articles. Note the pore (P). Proximal article (Px); distal article (Ds). Fig. 2. Basal region of a flat-tipped sensillum. Note the tubular body (TB) in longitudinal section. Inset: distal portion of the sensillum in cross section, with dendrites in the middle (arrow). Fig. 3. Numerous sensilla (arrowheads) located on a cercus. Fig. 4. Sensilla on a urotergite (arrows). Figs. 5-9. Distal border of antennal articles in several species that lack flat-tipped sensilla but which may present other kinds of fine setae (arrows). Fig. 5. Siphlaenigma janae. Fig. 6. Centroptilum luteolum. Fig. 7. Cloeon dipterum. Fig. 8. Procloeon bifidum. Fig. 9. Pseudocentroptilum pennulatum. This content downloaded from 157.55.39.247 on Wed, 27 Apr 2016 06:14:50 UTC All use subject to http://about.jstor.org/terms Flat-tipped sensillum as microcharacter

Video and Electron Microscopy of Particle Feeding in Sandwich Cultures of the Hexactinellid Sponge, Rhabdocalyptus dawsoni

Abstract: Cultured fragments of the hexactinellid sponge, Rhabdocalyptus dawsoni, were observed feeding on 1.0 Jm latex beads. Using contrast-enhanced video microscopy of live cultures, the events preceding, during, and after phagocytosis were observed. Particle ingestion occurred almost exclusively in flagellated areas where the water currents generated by the flagella of collar bodies could trap and hold particles against sponge tissue. Transmission electron microscopy of cultures fixed after feeding on latex beads showed that collar bodies themselves do not participate in particle phagocytosis. Rather, it is the primary and, probably, the secondary reticula of the trabecular reticulum in the flagellated chamber wall which effect phagocytosis. The intact sponge's tissue organization appears admirably suited for particle capture, and these sponges are likely to be, at least in part, particle feeders in their natural habitat. Additional key words: Hexactinellida, Porifera, filter feeding, culture The Hexactinellida (Porifera, subphylum Symplasma) is an enigmatic marine sponge group which has received little attention in the past due to their seclusion in deep waters. However, recent advances exploiting the few shallow-water species available for study have begun to reveal various aspects of their tissue organization, cell biology, and feeding (e.g. Reiswig & Mehl 1991; Boury-Esnault & Vacelet 1994; Leys 1995; Leys & Mackie 1997). Feeding studies have been pursued using two species: Oopsacas minuta TOPSENT 1927, a Mediterranean cave species (Perez 1996), and Rhabdocalyptus dawsoni (LAMBE 1892) a Northeast Pacific fjord species (Wyeth et al. 1996). Investigation of sponge feeding has focused on demosponges (Porifera, subphylum Cellularia), which has established them as non-specific particle feeders. Particles are extracted from water pumped through the body wall by choanocytes. Most tissues of the demosponge body wall have been implicated in particle phagocytosis. In addition to choanocytes, whose collar microvilli strain the current driven by the flagella, exopinacocytes of the dermal surface and endopinacocytes of the incurrent canals have been shown to engulf particles (Schmidt 1970; Diaz 1979; Willenz & Van de Vyver 1982; Imsiecke 1993). Furthermore, investigations of pumping rates (e.g., Reiswig 1971), a E-mail: rwyeth@u.washington.edu feeding and clearance efficiencies (e.g., Reiswig 1975a; Frost 1980), and the demosponge pump design (Larsen & Riisg'ard 1994) have also been conducted. The assumption that hexactinellid and demosponge feeding are the same should not be made a priori, since the two groups differ in the details of tissue organization and structure. In hexactinellids, the dermal membrane, trabeculae of the body wall, and flagellated chamber walls are all formed by the syncytial trabecular reticulum (Mackie & Singla 1983; Reiswig & Mehl 1991; Leys 1999). The hexactinellid aquiferous system consists of broad, expansive incurrent and excurrent canals bounded by a network of trabeculae (Schulze 1887; Ijima 1904; Leys 1999), unlike the smaller diameter canals, bounded by endopinacocytes, found in the demosponges used for feeding studies (Schmidt 1970; Diaz 1979; Imsiecke 1993). The large and oblong hexactinellid flagellated chambers, 60 (Jm or more in maximal dimension, are serviced by proportionately far more prosopyles (Schulze 1887; Ijima 1904; Reiswig & Mehl 1991; Leys 1999) than the smaller (30 Im in diameter) spherical chambers in demosponges used in feeding studies (Reiswig 1975b; Diaz 1979; de Vos et al. 1991). In hexactinellids, the trabecular reticulum branches to create a double reticulum in the chamber wall. The primary reticulum forms the wall of the chamber and encloses the bases of collar bodies. Inside the chamber, the secondary reticulum, offset from the primary, fills the gaps between This content downloaded from 207.46.13.176 on Mon, 20 Jun 2016 07:22:20 UTC All use subject to http://about.jstor.org/terms Hexactinellid sponge feeding collars inside the chamber (Reiswig 1979; Reiswig & Mehl 1991; Boury-Esnault & Vacelet 1994; Leys 1999). Hexactinellid collar bodies, although bearing similar collars of microvilli, are more widely spaced than demosponge choanocytes, are enucleate, and form part of a partially connected network within the flagellated chamber walls (Reiswig 1979; Mackie & Singla 1983; Leys 1999). Furthermore, the 20 tim flagella in hexactinellids are roughly 2-3 times the length of their associated collars (Reiswig 1979; Reiswig & Mehl 1991; Boury-Esnault & Vacelet 1994), considerably smaller than the equivalent ratio in demosponges which may have 45 pLm flagella, 4-6 times the collar length (see Larsen & Riisgard 1994). Although individually none of these structural differences leaps out as a reason to suspect substantial differences between hexactinellid and demosponge feeding, when considered together it is quite possible they could lead to considerable differences in the details of feeding

The anatomy of the hadal limpet Macleaniella moskalevi (Gastropoda: Cocculinoidea)

Abstract: The anatomy of the hadal, cocculinid limpet Macleaniella moskalevi is described. This species has a small, symmetrical, cap-shaped shell with a prominent internal transverse septum. As in other cocculinid genera, broad oral lappets and epipodial tentacles are present. Vestigial eyes are also present, but lack pigment and are modified into the basitentacular gland. The unique aortic vessel and associated hemal gland are present. Mantle cavity morphology is characteristic of cocculinid taxa; the cavity contains a pseudoplicatid gill, hypobranchial gland, and shallow brood pouch. Adults are simultaneous hermaphrodites, possessing a gonad with distinct egg and sperm producing regions; the common gonoduct is glandular. The single receptaculum seminis is connected to the anterior gonoduct via an enclosed duct. An enlarged right cephalic tentacle is inferred to function as a copulatory organ. The digestive gland occupies that part of the visceral mass confined above the septum and opens by way of two ducts into the distal esophagus. The anterior digestive tract is characterized by the presence of extremely well-developed esophageal pouches, a cuticularized sublingual pouch, and an unpaired jaw. A subradular sense organ and salivary glands are lacking. M. moskalevi is remarkably unique among previously described cocculinids in the anatomy of the reproductive tract and of the organs associated with the mantle cavity. In particular, the conformation of the nervous system in this species is unknown among cocculinids described thus far. Additional key words: Mollusca, deep-sea Cocculinids are limpet-like gastropods that are found in deep-sea habitats worldwide and are known primarily from bathyal depths. Macleaniella moskalevi LEAL & HARASEWYCH 1999 and Fedikovella caymanensis (Moskalev 1976) remain the only cocculinid species to date that are known to inhabit the hadal zone. Apart from the genus Teuthirostria, which inhabits chitinous cephalopod beaks, cocculinids are exclusively associated with submerged wooden substrates. Despite this consistent and predictable substrate association, it remains unclear whether cocculinids obtain nourishment directly from their substrate or from microbes, detritus, or other food resources associated with the surfaces on which they are found. Ever since their discovery and first description (Dall 1882), cocculinid species have intrigued their investigators with unique combinations of features. These features have proven problematic for the taxonomy of cocculinid limpets because they conflict with the suites a Author for correspondence. E-mail: eestrong@ gwis2.circ.gwu.edu of characters thought to be diagnostic at higher taxonomic levels. Difficulty in determining the taxonomic placement of cocculinids has been exacerbated by the common practice of placing any deep-sea limpet with a modified archaeogastropod organization in the same taxonomic group. Thus, the phylogenetic affinities of the family have remained ambiguous, a fact reflected in their fluctuating taxonomic status. Based on his anatomical investigations of Cocculina, Thiele (1903), stressed the isolated position of this family, but inferred a close relationship to Neritopsina. Later, Thiele (1908) suggested a close relationship between Bathysciadium and Cocculinidae and erected the superfamily Cocculinoidea to include the families Cocculinidae (Cocculina), Lepetellidae (Bathsciadium, Lepetella), and Addisoniidae (Addisonia). This scheme was expanded with the description of material collected during the latter half of the nineteenth century, but remained largely unchanged for the next 80 years. By the 1980's the Cocculinoidea had grown to include nine families representing a variety of deep-sea limpet-like forms: Addisoniidae (Dall 1882), Bathypeltidae (Moskalev 1971), Bathyphytophilidae (MosThis content downloaded from 207.46.13.129 on Sat, 25 Jun 2016 06:16:26 UTC All use subject to http://about.jstor.org/terms