Ciliary sieving and active ciliary response in capture of particles by suspension-feeding brachiopod larvae

Larvae of a brachiopod, Glottidia pyramidata, used at least two ciliary mechanisms to capture algal cells upstream from the lateral band of cilia that produces a feeding/swimming current. (1) Filtration: the larvae retained algal cells on the upstream (frontal) side of a sieve composed of a row of stationary laterofrontal cilia. Movement of the laterofrontal cilia could not be observed during capture or rejection of particles, but the laterofrontal cilia can bend toward the beating lateral cilia, a possible mechanism for releasing rejected particles from the ciliary sieve. (2) Localized changes of ciliary beat: the larvae may also concentrate particles by a local change in beat of lateral cilia in response to particles. The evidence is that the beat of lateral cilia changed coincident with captures of algal cells and that captured particles moved on paths consistent with a current redirected toward the frontal side of the tentacle by an induced local reversal of the lateral cilia. The change of beat of lateral cilia could have been an arrest rather than a reversal of ciliary beat, however. The similar ciliary bands in adult and larval lophophorates (brachiopods, phoronids, and bryozoans) suggest that these animals share a range of ciliary behaviours. The divergent accounts of ciliary feeding of lophophorates could be mostly the result of different authors observing different aspects of ciliary feeding.     

Ciliary filter-feeding structures in adult and larval gymnolaemate bryozoans

Abstract. Ciliary filter-feeding structures of gymnolaemate bryozoans—adults of Flustrellidra hispida and Alcyonidium gelatinosum , larvae of Membranipora sp.—were studied with SEM. In F. hispida and A. gelatinosum , the distal part of each tentacle has a straight row of stiff laterofrontal cilia which carry out "ciliary sieving" to capture suspended food particles that are subsequently transported downward towards the mouth by tentacle flicking; both structure and function resemble those of stenolaemate tentacles. The proximal part of the tentacle and of the ciliary ridge of a cyphonautes larva have strikingly similar structures, except that the laterofrontal cells are monociliate in the adults and biciliate in the larvae. The laterofrontal cells of the tentacles are arranged in a zigzag row and their cilia form two parallel rows, a frontal and a lateral row. The latter probably forms the sieve of stiff filter cilia in front of the water-pumping lateral cilia, whereas the frontal row appears to be held close to the frontal ciliary band of the tentacle. The biciliate laterofrontal cells of the cyphonautes larva have the cilia arranged in similar rows. The detailed morphological similarities between the ciliary bands of adult and larval filtering structures suggest that the feeding mechanisms are similar, contrary to what has been previously thought.     

Ciliary feeding structures and particle capture mechanism in the freshwater bryozoan Plumatella repens (Phylactolaemata)

Abstract. In contrast to marine bryozoans, the lophophore structure and the ciliary filter‐feeding mechanism in freshwater bryozoans have so far been only poorly described. Specimens of the phylactolaemate bryozoan Plumatella repens were studied to clarify the tentacular ciliary structures and the particle capture mechanism. Scanning electron microscopy revealed that the tentacles of the lophophore have a frontal band of densely packed cilia, and on each side a zigzag row of laterofrontal cilia and a band of lateral cilia. Phalloidin‐linked fluorescent dye showed no sign of muscular tissue within the tentacles. Video microscopy was used to describe basic characteristics of particle capture. Suspended particles in the incoming water flow, set up by the lateral ‘pump’ cilia on the tentacles, approach the tentacles with a velocity of 1–2 mm s^‐1. Near the tentacles, the particles are stopped by the stiff sensory laterofrontal cilia acting as a mechanical sieve, as previously seen in marine bryozoans. The particle capture mechanism suggested is based on the assumed ability of the sensory stiff laterofrontal cilia to be triggered by the deflection caused by the drag force of the through‐flowing water on a captured food particle. Thus, when a particle is stopped by the laterofrontal cilia, the otherwise stiff cilia are presumably triggered to make an inward flick which brings the restrained particle back into the downward directed main current, possibly to be captured again further down in the lophophore before being carried to the mouth via the food groove. No tentacle flicks and no transport of captured particles on the frontal side of the tentacles were observed. The velocity of the metachronal wave of the water‐pumping lateral cilia was measured to be ～0.2 mm s^‐1, the wavelength was ～7 μm, and hence the ciliary beat frequency estimated to be ～30 Hz (～20 °C). The filter feeding process in P. repens reported here resembles the ciliary sieving process described for marine bryozoans in recent years, although no tentacle flicks were observed in P. repens. The phylogenetic position of the phylactolaemates is discussed in the light of these findings.     

Mechanical sensitivity and cell coupling in the ciliated epithelial cells ofMytilus edulis gill

Summary Transmission electron microscopy has not provided strong evidence for gap junctions inMytilus edulis gill tissue, in spite of extensive physiological evidence for coupled ciliary arrest in lateral cells and coupled activation in abfrontal cells. To investigate the kinds and relative distribution of cell junctions and also to determine whether ciliary membrane particle differences exist in these two types of oppositely mechanically sensitive cells, we analyzed the structure of these and two other ciliated cell types (frontal and laterofrontal) by freeze-fracture replication. Gap junctions occur in all four ciliated cell types, but they are relatively small and of variable morphology, often consisting of elongate, winding complexes of membrane particles. Statistically, such structures rarely would be recognized as gap junctions in thin sections. Gap junctions appear to be most abundant between the highly coupled abfrontal cells, minimal between laterofrontal cells, and not evident in the epithelial cells that separate coupled ciliated cell types. The ciliary necklaces of the mechanically activated abfrontal cilia are typically 4- or 5-stranded while those of the remaining three cell types are mainly 3-stranded. In developing gill tips, ciliated cells have abundant gap junctions and newly formed cilia have a full complement of necklace particles. Nascent lateral cilia are not mechanically sensitive, indicating that the acquisition of mechanosensitivity does not correlate with the presence of ciliary necklace or other membrane particles. Lateral and laterofrontal cells become sensitive to neurotransmitters soon after the appearance of the latter during development, but mechanosensitivity of both lateral and abfrontal cells arises substantially later.     

The mussel filter–pump – present understanding,with a re‐examination of gill preparations

Filter feeding in mussels is a secondary adaptation where the gills have become W‐shaped and greatly enlarged, acting as the mussel filter–pump. Water pumping and particle capture in the blue mussel, Mytilus edulis, have been studied over many years. Here, we give a short status of the present understanding of ciliary structure and function of the mussel filter–pump, supplemented with new photo‐microscope and scanning electron microscopy (SEM) pictures of gill preparations. Pumping rate (filtration) and pressure to maintain flow have been extensively studied so the power delivered by the mussel pump to the water flow is known (1.1% of total respiratory power), but the actual cost based on gill respiration is much higher (19%), implying that the cost of maintaining of the large gill pump is considerable and that only relatively little energy can be saved by stopping or reducing the activity of the water‐pumping cilia so that continuous feeding with a ‘minimal scaled’ pump is cheaper than discontinuous feeding with a correspondingly larger pump. According to the present view, the pump proper is the beating lateral cilia (lc) on the gill filaments and particle capture is accomplished by the action of laterofrontal cirri (lfc) transferring particles from the main water current to the frontal gill filament currents driven by frontal cilia (fc). Unexplained aspects include retention efficiency according to particle size and the role of pro‐laterofrontal cilia (p‐lfc) placed between the lfc and fc. The structure of cilia and the mode of ciliary beating have been re‐examined in this study by new high‐resolution light and scanning electron microscopy of isolated gill preparations exposed to serotonin (5‐HT) stimulation which can activate the lc and lfc at low concentrations (10^?6 M), but removes the lfc from the interfilament canals at higher concentrations (10^?5 M).     

Contrasting scaling of ciliary filters in swimming larvae and sessile adults of fan worms (Annelida: Polychaeta)

Abstract. Both larval and adult fan worms capture particles with opposed bands of cilia. While the larvae use one of the opposed bands (the prototroch) for both feeding and swimming, the sessile adults rely partly on ambient currents to bring food particles to the ciliary bands. The scaling of length of prototrochal cilia with larval body size contrasts with scaling of the opposed latero-frontal cilia with adult body size. In the larva of the serpulid Hydroides elegans , the length of prototrochal cilia increased from 28 to 42 μm in early to late-stage larvae. In contrast, latero-frontal cilia did not increase in length (23 μm) during postlarval development of H. elegans. Among adults of 5 fan-worm species, lengths of latero-frontal cilia ranged from 22 to 35 μm and were weakly correlated with body size. The total area of ciliary filter nevertheless increased with increasing body dry weight of worms with an allometric exponent similar to exponents reported for gill and lophophore areas vs. body weight within species of suspension-feeding bivalves, brachiopods, and gastropods. The similar scaling was remarkable given the striking differences in distribution and function of the ciliary filters. In adult fan worms, increases in filter area depended largely on increases in number and length of radioles; differences in branching of radioles had little effect. Radioles were commonly in 2 or more rows in series, implying refiltration in still water by downstream radioles. Since the allometry of worms' filter area with body size depends on filters in series, it depends on ambient currents that overwhelm ciliary currents.     

Cinefilms of particle capture by an induced local change of beat of lateral cilia of a bryozoan

A local disruption of the metachronal wave always accompanies capture of algal cells by tentacles of Flustrellidra hispida (Fabricius). Beat changes for ≈0.2 s over ≈100μm of the ciliated band during capture of a 10-μm particle. The halted parcel of water is therefore larger than the particle of food but much smaller than the flow that continues past the tentacles elsewhere. These events are consistent with the hypothesis that an induced local reversal of beat concentrates particles for those suspension feeders that retain particles upstream from a band of simple cilia (adults or larvae of bryozoans, brachiopods, phoronids, hemichordates, and echinoderms). These events are not explained by other hypotheses that have been advanced for concentration of particles by these suspension feeders. Aerosol filtration models of direct interception are not applicable to this type of ciliary suspension feeder because retention depends on the magnitude of a stimulus and response to it. The stimulus will not be the same function of diameter of the food particle, and response is unlikely below a threshold stimulus.     

Putative sensory structures in marine bryozoans

Abstract. SEM studies of 21 species of marine bryozoans demonstrated that the abfrontal side of the tentacles bears a row of mono- or multiciliated cells, which are presumably sensory. In stenolaemates, the abfrontal cells, as well as the cells at the tentacle tips and the laterofrontal cells, are monociliated. In the 17 gymnolaemate species studied, each tentacle tip bears at least 3 multiciliated cells, each with a tuft of 5–7 stiff cilia of various lengths. On the abfrontal tentacle surface, mono- and multiciliated cells alternate, but all species studied have multiciliated cells at the base and the tip of each tentacle. In live animals, single cilia perform occasional flicks, whereas the tufts of 7–15 cilia on the multiciliated cells are immotile. Length and number of abfrontal cilia vary between species. Two types of multiciliated, putative sensory organs were found on the introvert of some gymnolaemates. One has an apical knob surrounded by a ring of cilia; the other has an apical tuft of cilia. The ultrastructure of the sensory cells of tentacles and introvert was studied in Rhamphostomella ovata . Our observations on both fixed and living material all suggest that these cells are primitive mechanoreceptors. The few species lacking ciliary structures on the introvert have long proximal ciliary tufts on the abfrontal tentacle surface.     

Opposed ciliary bands in the feeding larvae of sabellariid annelids

The larvae of marine annelids capture food using an unusual diversity of suspension-feeding mechanisms. Many of the feeding mechanisms of larval annelids are poorly known despite the abundance and ecological significance of both larvae and adults of some annelid taxa. Here we show that larvae of two species of sabellariid annelids, Sabellaria cementarium and Phragmatopoma californica, bear prototrochal and metatrochal cilia that beat in opposition to each other. For larvae of S. cementarium, we provide evidence that these opposed bands of cilia are used to capture suspended particles. In video recordings, captured particles were overtaken by a prototrochal cilium and then moved with the cilium to the food groove, a band of cilia between the prototroch and metatroch. They were then transported by cilia of the food groove to the mouth. Lengths of the prototrochal cilia, lengths of the prototrochal ciliary band, size range of the particles captured, and estimated rates of clearance increased with larval age and body size. Confirmation of the presence of opposed bands in larvae of sabellariids extends their known occurrence in the annelids to members of 10 families. Opposed bands in these different taxa differ in the arrangements and spacing of prototrochal and metatrochal cilia, and in whether they are used in combination with other feeding mechanisms. Opposed bands appear to be particularly widespread among the larvae of sabellidan annelids (a clade that includes sabellariids, sabellids, and serpulids), even in some species whose larvae do not feed. A parsimony analysis suggests that opposed bands are ancestral in this clade of annelids.     

Numerical simulation of flow fields and particle trajectories in ciliary suspension feeding

A model describing the ciliary driven flow and motion of suspended particles in downstream suspension feeders is developed. The quasi-steady Stokes equations for creeping flow are solved numerically in an unbounded fluid domain around cylindrical bodies using a boundary integral formulation. The time-dependent flow is approximated with a continuous sequence of steady state creeping flow fields, where metachronously beating ciliary bands are modelled by linear combinations of singularity solutions to the Stokes equations. Generally, the computed flow fields can be divided into an unsteady region close to the driving ciliary bands and a steady region covering the remaining fluid domain. The size of the unsteady region appears to be comparable to the metachronal wavelength of the ciliary band. A systematic investigation is performed of trajectories of infinitely small (fluid) particles in the simulated unsteady ciliary driven flow. A fraction of particles appear to follow trajectories, that resemble experimentally observed particle capture events in the downstream feeding system of the polycheate Sabella penicillus, indicating that particles can be captured by ciliary systems without mechanical contact between particle and cilia. A local capture efficiency is defined and its value computed for various values of beat frequencies and other parameters. The results indicate that the simulated particle capture process is most effective when the flow field oscillates within timescales comparable to transit timescales of suspended particles passing the unsteady region near the ciliary bands. However, the computed retention efficiencies are found to be much lower than those obtained experimentally.     

Studies on the gill ciliation of the American oyster Crassostrea virginica (Gmelin)

Latero-frontal, para-latero-frontal, and frontal ciliary tracts on the gill filaments of Crassostrea virginica (Gmelin) were studied with light microscopy and scanning electron microscopy. Latero-frontal cirri are complex structures composed of varying numbers of paired cilia. The multiple pairs of cilia which constitute a single cirrus are closely appressed for a portion of their length; they then branch laterally from the central axis in a plume-like fashion. Latero-frontal cirri of adjacent gill filaments create a filtration sieve which should be capable of retaining particles smaller than 1 μm in diameter. Para-latero-frontal cilia are short, closely spaced cilia arranged as a staggered row along the frontal side of each tract of latero-frontal cirri. Latero-frontal cirri and para-latero-frontal cilia occur on ordinary, principal, and transitional gill filaments. Frontal ciliary tracts of ordinary filaments are divided into a central, ventrally directed coarse tract, flanked on either side by a dorsally directed fine ciliary tract. The coarse tract is covered by cirri which are comprised of five to eight cilia, while the fine frontal tracts are made up of individually functioning cilia. The frontal ciliary tracts of principal and transitional filaments bear only dorsally directed fine cilia. The unique direction of effective beat of the coarse frontal cirri of ordinary filaments, in combination with the action of fine frontal cilia and the strategic location of mucus producing cells, is used to describe a possible mechanism for the sorting of filtered particles.     

Opposed bands of cilia in the nonfeeding larvae of the serpulid annelid Salmacina tribranchiata

The nonfeeding planktonic larvae of marine invertebrates typically lack larval feeding structures. One puzzling exception to this generalization is the annelid clade Sabellidae, in which nonfeeding larvae possess ciliary bands (specifically, food groove and metatroch) that, to the best of our knowledge, have no function other than in feeding. Nishi and Yamasu (1992b, Bulletin of the College of Sciences, University of the Ryukyus, 54 , 107–121) published a scanning electron micrograph showing that nonfeeding larvae of the serpulid annelid Salmacina dysteri also possess food groove and metatrochal cilia. Here I demonstrate that nonfeeding larvae of Salmacina tribranchiata also bear ciliary bands identifiable as food groove and metatroch by position. High‐speed video of ciliary beat patterns shows that, together with the prototrochal cilia, these bands function in an opposed band system. The presence of feeding structures in nonfeeding annelid larvae is thus more widely distributed than previously recognized. The presence of feeding structures may make evolutionary transitions to planktotrophy more likely, and may underlie an inferred origin of larval feeding in the common ancestor of one of the two major clades of serpulid annelids, Serpulinae.     

ICK is essential for cell type‐specific ciliogenesis and the regulation of ciliary transport

Cilia and flagella are formed and maintained by intraflagellar transport (IFT) and play important roles in sensing and moving across species. At the distal tip of the cilia/flagella, IFT complexes turn around to switch from anterograde to retrograde transport; however, the underlying regulatory mechanism is unclear. Here, we identified ICK localization at the tip of cilia as a regulator of ciliary transport. In ICK‐deficient mice, we found ciliary defects in neuronal progenitor cells with Hedgehog signal defects. ICK‐deficient cells formed cilia with mislocalized Hedgehog signaling components. Loss of ICK caused the accumulation of IFT‐A, IFT‐B, and BBSome components at the ciliary tips. In contrast, overexpression of ICK induced the strong accumulation of IFT‐B, but not IFT‐A or BBSome components at ciliary tips. In addition, ICK directly phosphorylated Kif3a, while inhibition of this Kif3a phosphorylation affected ciliary formation. Our results suggest that ICK is a Kif3a kinase and essential for proper ciliogenesis in development by regulating ciliary transport at the tip of cilia.     

Swimming Speed of Larval Snail Does Not Correlate with Size and Ciliary Beat Frequency

Many marine invertebrates have planktonic larvae with cilia used for both propulsion and capturing of food particles. Hence, changes in ciliary activity have implications for larval nutrition and ability to navigate the water column, which in turn affect survival and dispersal. Using high-speed high-resolution microvideography, we examined the relationship between swimming speed, velar arrangements, and ciliary beat frequency of freely swimming veliger larvae of the gastropod Crepidula fornicata over the course of larval development. Average swimming speed was greatest 6 days post hatching, suggesting a reduction in swimming speed towards settlement. At a given age, veliger larvae have highly variable speeds (0.8–4 body lengths s⁻¹) that are independent of shell size. Contrary to the hypothesis that an increase in ciliary beat frequency increases work done, and therefore speed, there was no significant correlation between swimming speed and ciliary beat frequency. Instead, there are significant correlations between swimming speed and visible area of the velar lobe, and distance between centroids of velum and larval shell. These observations suggest an alternative hypothesis that, instead of modifying ciliary beat frequency, larval C. fornicata modify swimming through adjustment of velum extension or orientation. The ability to adjust velum position could influence particle capture efficiency and fluid disturbance and help promote survival in the plankton.     

Cilia-Mediated Oriented Chemokinesis in Tetrahymena thermophila

The role of the cilia in the locomotion (“gliding”) of Tetrahymena thermophila in a semi-solid medium has been studied when cells were migrating in gradients of attractant. Video recordings and computer-aided motion analysis of migrating cells and their ciliary activity show that Tetrahymena thermophila migrate by swimming forward in semi-solid methyl cellulose, using their cilia. Ciliary reversals occur at certain intervals and cause a termination (“stop”) of cellular migration. Cells with reversed cilia resume forward migration when normal ciliary beating resumes. In gradients of attractants, cells migrating towards the attractant suppress ciliary reversals, which leads to longer runs between stops than in control cells. Cells migrating away from the attractant have a higher frequency of ciliary reversals than the control cells resulting in shorter runs. Stimulated cells adapt to a particular ambient concentration of attractant several times during migration in the gradient. Adaptation is followed by de-adaptation, which occurs during the “stop”. In the presence of cycloheximide, a strong inhibitor of chemoattraction, the attractant-induced suppression of ciliary reversal is abolished (cells become desensitized to the attractant). It is concluded that Tetrahymena has a short-term memory during adaptation. This is important for the efficiency of migration towards an attractant.     

Calcium control of ciliary arrest in mussel gill cells.

After several hours in 20 mM sodium phosphate and 40 mM KCI (pH 7.4) or similar simple solutions, ciliated cells exfoliate en masse from stripped gill epithelium of freshwater mussels, e.g., Elliptio complanatus. Three types of ciliated cells--lateral (L), laterofrontal (LF), and frontal (F)--can be distiniguished and counted separately in the suspensions. About one-half of the cells of each type remain motile. Motility is unaffected by addition of 10(-5) M A23187 or 10(-2) M Ca+2 added separately, but when ionophore and Ca+2 are added together, ciliary beat is largely arrested. Treatment of the cells with Triton X-100 (Rohm & Haas Co., Philadelphia, Pa.) results in a total loss of motility as the ciliary membrane becomes disrupted. Such models can be reactivated by addition of ATP and Mg+2. All ciliated cell types are reactivated to about the same extent. At least 80% of the activity of the untreated preparation returns. Ca+2-EGTA buffers added to the reactivating solutions permit titration of free Ca+2 concentration vs. percent motility. Activity is unchanged for all cell types at Ca+2 less than 10(-7) M; at 10(-6) Ca+2, L cilia of all cell types are arrested differentially, whereas at Ca+2 greater than 10(-4) M most cilia of all cell types are arrested. We conclude: (a) that increasing cytoplasmic Ca+2 is directly responsible for ciliary arrest, (b) that the readily reversible physiological arrest response of the L cilia in the intact gill is caused by a rise in free Ca+2 in narrow limits from ca. 5 x 10(-7) M to ca. 8 x 10(-7) M, and (c) that the site which is sensitive to Ca+2 is part of the ciliary axoneme or the basal apparatus.     

The ciliary photoreceptor in the teleplanic veliger larvae of Smaragdia sp. and Strombus sp. (Mollusca,Gastropoda)

The cerebrally innervated eyes of the veliger larvae of Smaragdia sp. and Strombus sp. are composed of a lens, a cornea, and an everse retina. The retina contains two different types of cells, ciliary sensory cells and supportive cells which bear one or two cilia. It is suggested that: (a) the ciliary photoreceptors of these teleplanic veliger larvae are correlated with a long pelagic life in the ocean, which can last up to twelve months, and (b) that structural details of the photoreceptors can change during ontogenesis (ciliary vs rhabdomeric). Furthermore, the cilia of the supportive cells apparently tranport lens material and thus play an important role in lens formation. A decomposition mechanism of pigment granules is examined.Abbreviations bb basal body - bp basal plate - c cilium - cc corneal cell - cm ciliary membranes - cw ciliary whorl - ecm extracellular matrix - gr electron-dense granules - l lens - lb lamellar body - mp membranous pieces - mt microtubules - mv microvilli - n nucleus - oc optic cavity - on optic nerve - pg pigment granule - sc sensory cell - sj septate junction - spc supportive cell - v vesicles     

Ciliated band structure in planktotrophic and lecithotrophic larvae of Heliocidaris species (Echinodermata: Echinoidea): a demonstration of conservation and change

The evolution of lecithotrophic (non-feeding) development in sea urchins is associated with reduction or loss of structures found in the planktotrophic (feeding) echinopluteus larvae. Reductions or losses of larval feeding structures include pluteal arms, their supporting skeleton and the ciliated band that borders them. The barrel-shaped lecithotrophic larva of Heliocidaris erythrogramma has, at its posterior end, two or three ciliated band segments comprised of densely packed, elongate cilia. These cilia may be expressions of the epaulettes that would have been present in an ancestral larval form, represented today by the feeding echinopluteus of H. tuberculata . We compared the development and cellular organization of the larval ciliary structures of both Heliocidaris species to assess whether the ciliary bands of H. erythrogramma are expressions of the feeding ciliated band or epaulettes of an echinopluteus. Epaulette development in feeding larvae of H. tuberculata involves separation of specific parts of the ciliated band from the rest of the feeding ciliated band, hyperplastic addition of ciliated cells and hypertrophic growth of the cilia. Like epaulettes, the ciliated bands of H. erythrogramma are composed of long spindle-shaped cells arranged in a cup-shaped collection that bulges into the blastocoel; and these cells have elongated cilia. In their developmental origin and topological arrangement however, the ciliated bands of H. erythrogramma correspond more closely with parts of the pluteal feeding ciliated band than with epaulettes. The larvae of this echinoid appear to develop epaulette-like bands from parts of the original (but reduced) feeding ciliated band. The evolution of development in H. erythrogramma has thus involved both conservation and change in echinopluteal ciliary structures.     

Scanning Electron Microscopy of the Adoral Ciliary Zone of Entodinium Stein (Ciliophora,Entodiniomorphida)1

The adoral ciliary zone of the rumen ciliates, Entodinium spp., was observed topographically in the SEM. The upper side of the anterior end of the body was indented by the vestibulum, which had cilia arranged on its right wall and ribs along the left wall. The adoral ciliary zone could be divided into at least two arrangements. The outer ciliary zone had many membranelle-like structures, which consisted of cilia arranged radially from the body axis. Each membranelle-like structure consisted of two rows of about eight cilia each lining up in a single file. It was, however, different from a typical membranelle, because its cilia were connected with the vestibular cilia and were arranged not spirally but on a plane. These cilia extended toward the outside because of the projecting cytoplasm from which they originated. In contrast, the cilia of the inner ciliary zone were aggregated to form relatively unsystematic bundles. Since the vestibular opening was slanted on the upper side of the body, the ciliary bundles were thick on the lower side and sparse on the upper side of the body. Neither outer nor inner ciliary zones completely surrounded the vestibular opening. The ciliature started from the left side of the vestibular opening, encircled the lower side of the body, and entered the vestibulum from its right side. The functions of these two types of ciliary arrangements are discussed.     

The role of hair cells, cilia and ciliary motility in otolith formation in the zebrafish otic vesicle

Otoliths are biomineralised structures required for the sensation of gravity, linear acceleration and sound in the zebrafish ear. Otolith precursor particles, initially distributed throughout the otic vesicle lumen, become tethered to the tips of hair cell kinocilia (tether cilia) at the otic vesicle poles, forming two otoliths. We have used high-speed video microscopy to investigate the role of cilia and ciliary motility in otolith formation. In wild-type ears, groups of motile cilia are present at the otic vesicle poles, surrounding the immotile tether cilia. A few motile cilia are also found on the medial wall, but most cilia (92-98%) in the otic vesicle are immotile. In mutants with defective cilia (iguana) or ciliary motility (lrrc50), otoliths are frequently ectopic, untethered or fused. Nevertheless, neither cilia nor ciliary motility are absolutely required for otolith tethering: a mutant that lacks cilia completely (MZovl) is still capable of tethering otoliths at the otic vesicle poles. In embryos with attenuated Notch signalling [mindbomb mutant or Su(H) morphant], supernumerary hair cells develop and otolith precursor particles bind to the tips of all kinocilia, or bind directly to the hair cells' apical surface if cilia are absent [MZovl injected with a Su(H)1+2 morpholino]. However, if the first hair cells are missing (atoh1b morphant), otolith formation is severely disrupted and delayed. Our data support a model in which hair cells produce an otolith precursor-binding factor, normally localised to tether cell kinocilia. We also show that embryonic movement plays a minor role in the formation of normal otoliths.