Opposed ciliary bands in the feeding larvae of sabellariid annelids.
Studies of the functional morphology of suspension feeding by marine invertebrate larvae are of interest in a number of ecological and evolutionary contexts. Knowledge of how particular larval morphologies and feeding mechanisms affect feeding performance can, for example, be used to make inferences on the growth and survivorship of larvae in different environments (e.g., McEdward and Strathmann, 1987; Stone, 1989; Vargas et al 2006). At another level, comparative data on the diversity of larval feeding mechanisms in marine invertebrates are essential in evaluating hypotheses on how larval nutritional mode itself has evolved, a question that has received substantial attention in the past few decades (e.g., Strathmann, 1978; Rouse, 2000). In the absence of fossil evidence, reconstructions of ancestral character states depend on the distribution of traits in existing animals, their inferred relationships, and a model of transitions between character states (Keever and Hart, 2008). Studies of larval feeding mechanisms extend the breadth of character sampling in extant animals. In some cases, observations of larval feeding also contribute to models of transitions between character states (Hadfield et al., 1997; Miner et al., 1999).
The Annelida is an especially interesting clade in which to examine both of these aspects of larval feeding, largely because of the impressive diversity of larval forms that it contains. Both nonfeeding and feeding larval development are common in the clade (Pernet et al, 2002; Rouse, 2006). Among the species with feeding larval stages, larval feeding mechanisms are unusually varied, especially as compared to more intensively studied clades such as the echinoderms and molluscs, where larvae of most or all species appear to feed using similar mechanisms (R. R. Strathmann, 1987; Hart and Strathmann, 1995). Variation among annelids in larval feeding mechanisms makes possible close comparisons of the performance consequences of those different mechanisms (Phillips and Pernet, 1996). Further, information on the functional morphology of larval feeding in annelids may strongly affect inferences on how larval nutritional mode and feeding mechanisms have evolved in the clade.
A prerequisite for such analyses, however, is information on the morphology and behavior associated with these diverse feeding mechanisms. Such data are surprisingly sparse for the Annelida. At least 22 families of annelids include species that have feeding larvae (see table 6.1 in Rouse, 2006); for larvae in many of these families, feeding mechanisms are either poorly understood or completely uncharacterized. This is true even for the larvae of species whose adult forms are ecologically important, or whose larvae are common in coastal plankton (e.g., polychaetes in the families Spionidae, Nephtyidae, or Sabellariidae). The paucity of these data hampers attempts to understand the ecological consequences of diverse larval feeding mechanisms or to understand how larval nutritional modes and feeding mechanisms have evolved in annelids and members of related phyla.
Here we provide new data on the functional morphology and performance of feeding in larvae of the annelid family Sabellariidae. Adult sabellariids construct and inhabit tubes of sand grains. Members of a few species settle in aggregations and produce massive intertidal or subtidal reefs. Their larvae are common members of coastal plankton communities in many locations, and the larvae of all studied species require particulate food while in the plankton (Giangrande, 1997).
On the basis of unpublished observations of particle capture, R. R. Strathmann (1987) concluded that larvae of Sabellaria alveolata captured particles between opposed bands of cilia. This feeding mechanism, in which a preoral band of compound cilia (the prototroch) beats from anterior to posterior, and a postoral band of compound cilia (the metatroch) beats in the opposite direction, trapping particles between them, was described in the larvae of serpulid annelids (Strathmann et al., 1972), close relatives of the sa-bellariids (Kupriyanova and Rouse, 2008). Larval feeding using opposed bands of cilia is also known in members of several other familes of annelids (Pernet et al., 2002; Pernet and McArthur, 2006).
Nevertheless, how the larvae of sabellariids capture particulate food has remained controversial, in part because the many published studies of the morphology of sabellariid larvae (e.g., Wilson, 1929, 1968; Dales, 1952; Cazaux, 1964; Smith and Chia, 1985) are ambiguous as to whether two of the ciliary bands necessary for this feeding mecha-nism--the metatroch and food groove--are even present in sabellariid larvae. None of these authors mentioned the presence of metatroch or food groove cilia in the studied larvae, but none of these authors indicated that they looked for them, either. We assume that the lack of interest in metatroch and food groove ciliary bands in the earlier studies is because their functional significance was not well understood prior to the early 1970s. However, this omission is still somewhat surprising, as these ciliary bands are not particularly difficult to see with compound microscopy. Indeed, cilia of the metatroch and food groove were illustrated in larvae of several other annelid families in the nineteenth century by Hatschek (1878, 1880, 1885). Further, what we interpret as metatrochal cilia are shown (but not labeled or otherwise mentioned) in a low-magnification micrograph of the larva of S. alveolata published by D. P Wilson (1968, his fig. 1A, on the larva's left side, posterior to the prototroch).
[FIGURE 1 OMITTED]
Strathmann and Grunbaum (2006) later published a single still image clearly showing prototroch and metatrochal cilia in larvae of S. cementarium, but a detailed description of larval feeding in members of this family is still not available. This is unfortunate, as such a description would be useful in attempts to understand the evolution of larval nutritional mode in the Sabellida, the clade of annelids that includes the sabellariids and whose phylogenetic relationships are rapidly becoming better known (e.g., Lehrke et al., 2007; Kupriyanova and Rouse, 2008). In this paper, we more fully describe the ciliary bands of larvae of the sabellariids S. cementarium Moore 1906 and Phragmatomopa californica Kinberg 1867. Previous detailed descriptions of the morphology of larvae of the these species do not mention the presence of metatroch or food groove ciliary bands (Dales, 1952; Smith and Chia, 1985). We also report on direct observations of particle capture and transport by opposed bands of cilia in larvae of S. cementarium, and on the feeding performance of larvae of S. cementarium as a function of age and size.
Materials and Methods
Collection of adults, spawning, and larval culture
Small clumps of adult worms were collected by dredge at various sites around the San Juan Islands, Washington (Sabellaria cementarium), or by hand in the intertidal zone in southern California (Phragmatopoma californica). Adults were maintained without supplemental food in tanks of flowing seawater for days to months before use. To obtain eggs and sperm, individual worms were removed from their sand tubes and isolated in small dishes of filtered seawater. Gametes were typically released from segmental nephridiopores within a few minutes of removal from the tube. Oocytes from several females were washed in coarsely filtered (5-[micro]m-mesh size) seawater, then fertilized with a dilute suspension of sperm from several males. Embryos and larvae were cultured in glass jars kept at local surface seawater temperatures ([approximately equal to]10-12 [degrees]C) and stirred with a swinging paddle apparatus (M. F. Strathmann, 1987). Larvae were cultured at densities of [approximately equal to]1-5 larvae [ml.sup.-1]; once or twice a week, culture water was partially changed by reverse filtration and replacement with filtered seawater. After water changes, larvae were fed a mixture of cells of Isochrysis sp. and Rhodomonas sp. Algal concentrations were not measured, but were high and unlikely to be limiting for larval growth.
Larval morphology and feeding behavior
The ciliary bands of larvae were examined with light and scanning electron microscopy (SEM). For SEM, larvae were relaxed by incubation in a 1:1 mixture of seawater and 7.5% [MgCl.sub.2], then fixed in 2% [OsO.sub.4] in 1.25% sodium bicarbonate buffer (pH 7.2) at 5 [degrees]C for 1-3 h. They were then rinsed in distilled water, dehydrated through an ascending concentration series of ethanol, critical-point dried, and mounted on stubs with carbon adhesive disks. Larvae were viewed and photographed with a JEOL JSM-35 scanning electron microscope.
For larvae of S. cementarium, beat patterns of feeding cilia and particle capture and transport were recorded with a high-speed video camera (Motionscope 1000-S, RedLake Imaging Inc.) mounted on a compound microscope with DIC optics. Larvae were mounted on slides. Some were lightly pinned beneath a coverslip supported with plasticene (modeling clay) at its corners. Some were attached by cyanoacrylic glue to the thin part of a glass needle that had been pulled from a glass rod held in a gas flame. The larva and part of the needle were in seawater beneath a coverglass supported at the corners with plasticene. The thicker part of the needle was mounted on the slide with plasticene beyond the edge of the coverglass so that the needle could be rotated to change orientation of the larva. This arrangement tethered the larva while providing more space for ciliary currents. In some cases larvae were observed in the absence of food particles, but in other cases algal cells or polystyrene beads were introduced at the edge of the coverslip. Images were collected at 125-500 frames per second and replayed at 10 frames per second. Sequences were routed from the camera through an analog-digital converter to a Macintosh computer, and saved as iMovie files (Apple Computer Inc.) for later examination.
Velocities of cilia and particles
We estimated velocities of the cilia of larvae of S. ce-mentarium and surrounding particles from a video clip of 2.04 s with frames recorded at 8-ms intervals. The clip was selected for the orientation and activity of the tethered larva. Tethered larvae captured particles at low clearance rates. The effective strokes of the prototroch cilia and the paths of particles were often oblique to the plane of focus. While other clips, including those with 2-ms intervals, provided information on ciliary beat and captures, the 2-s clip provided the greatest frequency of captures, with four near the plane of focus, together with effective strokes close to the plane of focus. In the 2-s video clip, the particles were cells of the alga Isochrysis galbana. Estimates of angular velocities of cilia were approximate because the axis of rotation for an effective stroke could not always be identified accurately and because of changing curvature of cilia during an effective stroke. Estimates of angular velocities of particles were approximate because distance of particles from the axis of the effective stroke changed between 2% and 33% as they passed through the beating prototroch. Velocities of particles as distance per time were approximated from straight line distance over an 8-ms interval. With arcs up to 50[degrees], the maximum error was an underestimate of 3%. The microscope lamp elevated temperatures of larvae on the slide somewhat above the room temperature of about 18 [degrees]C.
To check for variation among tethered larvae, we estimated angular velocities of prototrochal cilia and ratios of velocities of cilium and particle for four individuals. One of the larvae was the specimen selected for more detailed analysis. Measurements were for 10 particles passing through the prototrochal effective strokes of each larva, during a single 8-ms interval within the middle third of the effective stroke for each particle.
Larval feeding performance
We estimated clearance rates of larvae of three ages (9-, 21-, and 30-d post-fertilization) on polystyrene spheres of several sizes (3-45-[micro]m diameter). Clearance rates of larvae are usually estimated in one of three ways: from direct observations of capture rates by larvae swimming in suspensions of food particles of known concentration for short periods; from estimates of ingestion rates of larvae held in food suspensions for short periods; or from estimates of rates of depletion of food particles from suspensions over longer periods. Though the first method should give the best estimates of maximal clearance rates, it is difficult to use except with larvae that swim slowly (e.g., echinoderms: Hart, 1991) feeding on relatively large particles. We measured clearance rates of larvae of S. cementarium by the second method. Our results are underestimates of maximal rates because they were calculated as means of the ingestion rates of numerous larvae (some of which were probably not feeding at maximal rates during the test period), because not all particles captured were necessarily ingested, and because some ingested particles may have been defecated or regurgitated before being counted.
Oocytes obtained from several females were pooled and fertilized with sperm obtained from several males, then split into several jars and and cultured as described above. Larvae in this set of cultures developed roughly synchronously. Before each feeding trial, larvae haphazardly selected from the culture jars were isolated in filtered seawater (FSW) without food for 24 h to increase the likelihood that they would be feeding at maximal rates. At each feeding trial, 60 of these starved larvae were distributed among six 25-[micro]l glass vials filled with FSW (10 larvae per vial). Food particles of one of six sizes (polystyrene beads with diameters of 3, 6, 10, 15, 20, or 45 [micro]m--see below) were added to each vial to a final concentration of 2000 beads*m[l.sup.-1]. Vials were then strapped to a rotating plankton wheel (2.5 rpm) for 20 min, after which larvae were immediately killed by the addition of 0.8 ml of 100% formalin. As many larvae as could be recovered (7-10) from each vial were later examined at 400X, and the beads found in each larva's gut were enumerated. From these data the mean ingestion rate of larvae from a single vial was estimated. At each age, the experiment was replicated three times; the mean of the three replicate vials fed the same size particles was taken as an estimate of ingestion rate for larvae of that age on particles of that size. Clearance rates were calculated as ingestion rate divided by available particle concentration.
Spherical polystyrene beads (Polybeads, Polysciences Inc.) were used as "food" particles so that consequences of particle size on clearance rate could be examined independently of properties such as shape and flavor. Suspensions were prepared by rinsing beads once in distilled water, then resuspending them in 500 ul of 2.5% bovine serum albumin (BSA, Sigma A-3311) in distilled water. Beads were incubated in 2.5% BSA for 2 h at room temperature, rinsed once in FSW, and resuspended in FSW. Concentrations of these stock suspensions were estimated as the means of 3-4 replicate counts on a hemacytometer. Stock suspensions were used in experiments within 1 h of resuspension in FSW. Incubation in 2.5% BSA greatly reduced any clumping of beads. Further, preliminary experiments with the larvae of the annelid Hydroides sanctaecrucis (whose larvae are similar in size to those of S. cementarium and feed using a very similar mechanism) showed that BSA-incubated beads were ingested at rates similar to those of beads that were incubated in algal (Isochrysis and Nannochloris) exudates; both BSA-and algal exudate-incubated beads were ingested at higher rates than plain beads (Pernet, unpubl. data). We assume that this is also true in S. cementarium. Because transparent 3-and 6-[micro]m beads were difficult to count accurately in larval guts, for these two size classes we used blue-dyed polystyrene beads. In H. sanctaecrucis, blue-dyed 6-[micro]m beads were ingested at rates similar to plain 6-[micro]m beads (Pernet, unpubl. data); we assumed the same was true for 3-[micro]m beads and that these results held for S. cementarium as well.
The number of particles found in larval guts reflects ingestion rates only if larvae are eating at a constant rate during experiments, and if particles are not lost from guts by defecation during experiments. In preliminary experiments on larvae of H. sanctaecrucis, we examined ingestion rates of 6-and 10-[micro]m beads in high-concentration suspensions (5000 beads*m[l.sup.-1]) by 2-d-old larvae. Ingestion increased linearly with time up to at least 20 min, suggesting that larvae would not become satiated or begin to defecate within this time period (Pernet, unpubl. data). We chose to run our experiments for 20 min at lower concentrations of beads (2000 beads * [ml.sup.-1]). Further preliminary experiments on larvae of H. sanctaecrucis showed that clearance rates were constant over particle concentrations ranging from 500 to 5000 beads * [ml.sup.-1] (Pernet, unpubl. data).
At each feeding experiment, we measured the linear dimensions of 8-10 larvae per culture. We mounted live larvae in a drop of seawater, relaxed them with an equal volume of 7.5% [MgCl.sub.2], and slowed their movements by lightly dusting the drop with polyethylene oxide. We supported coverslips with clay at their corners so as not to compress the larvae. Each larva was briefly videotaped in lateral view with a 20X objective lens for measurements of body length and diameter. Larvae were videotaped with a 40X objective lens for measurements of prototrochal ciliary length. Videotaped images were later traced onto acetate sheets from a monitor. Three dimensions were measured from each larva: body length, body diameter at the level of the prototroch, and length of prototrochal cilia (Fig. 1). Ciliary length was estimated as the mean of measurements of 3-4 cilia from each larva. A videotaped image of a stage micrometer served to calibrate measurements.
Larval ciliary bands
Scanning electron micrographs of larvae of Sabellaria cementarium clearly demonstrated the presence of three equatorial ciliary bands located in the positions typical of an opposed-band system of particle capture--prototroch (preoral), metatroch (postoral), and food groove (circumoral) (Fig. 1A, B). These circled the body completely except for a substantial dorsal gap in all three ciliary bands (Fig. 1C), as previously described for the prototroch by Wilson (1929). Videos of beating cilia indicated that the cilia posterior to the mouth, in the same position as the metatroch, were longer than metatrochal cilia elsewhere. Other aspects of larval form were as previously described by Smith and Chia (1985). Larvae of Phragmatopoma californica also bore typical opposed bands of cilia (Fig. 1D), and they also had a dorsal gap in all three ciliary bands (as previously described for the prototroch by Dales, 1952). Such dorsal gaps in opposed bands are not unique to the larvae of sabellariids, being found in larvae of several other taxa that bear opposed bands (e.g., sabellids: Pernet, 2003; spionids: Pernet and McArthur, 2006).
Particle capture and transport
Video recordings of larvae of S. cementarium at 125 and 500 frames per second demonstrated that the effective strokes of prototroch and metatroch are toward each other and the recovery strokes are away from each other (Fig. 2). The longer cilia posterior to the mouth similarly beat in opposition to the prototrochal cilia. At times the metatroch cilia remained stationary while the prototroch cilia continued to beat.
[FIGURE 2 OMITTED]
Tethered larvae did not feed at high rates, in that many algal cells passed within reach of the prototrochal cilia without being captured; however, tethered larvae did capture some particles. Algal cells that entered the range of effective strokes of the prototrochal cilia were overtaken by the cilia, whether captured or not. In captures, the overtaken particle then moved with the cilium until it was near the food groove. In Figure 3, a cell of I. galbana was being overtaken at 0 ms and was with the cilium at 24, 32, and 40 ms. The cell was then moved anteriorly near the recovery strokes at 64 and 72 ms, before being moved around the food groove toward the mouth. A second cell of I. galbana that was within range of effective strokes but not yet contacted at 232 ms was in contact with a prototrochal cilium from 240 to 264 ms, then was left by the cilium at 288 ms (Fig. 3). At 312 to 336 ms this cell was moved in the food groove out of the focal plane, toward the mouth. The relation between a captured particle and individual prototrochal cilia was less clear for the particles (not shown) that passed through the prototroch close to the recovery strokes.
The video recordings provide estimates of paths and speeds of particles and cilia. The total arc of beat at the base of a prototrochal cilium exceeds 200[degrees] (Fig. 2). Particle paths and speeds were estimated in parts of the strokes between 54[degrees] and 140[degrees]. The mean length of prototrochal cilia, measured at the end of their effective stroke, was estimated to be 55.8 [micro]m (SD 1.3 [micro]m, n = 20). During the middle of the effective strokes, the moving tips blurred into invisibility. Even at the end of the effective stroke, the tips faded into invisibility, so that some inaccuracy in estimates of ciliary length is possible despite the apparent precision of measurements. In a 2-s recording with four captures of particles, the four captures were of particles passing between 17 and 27 [micro]m from the base of the prototrochal cilia, a distance measured when particles were between 80[degrees] and 120[degrees] of arc of the beat. Thus observed captures occurred within a small part of the length of prototrochal cilia and a small part of the ciliary current, but captures by tethered larvae may not be representative of the length of cilium that can capture a particle. Even within the 17-27-[micro]m zone, many particles that were close to prototrochal cilia were passed by cilia rather than traveling with cilia throughout their strokes toward the food groove.
During captures, cilia traveled with particles after overtaking them. For a similar part of the ciliary beat (88[degrees]-170[degrees]), cilia traveled faster than particles when passing them. Ratios of angular velocities of cilium to particle were 1.3 to 1.5 for particles being overtaken, left, or passed at 26-41 um from the base of the cilium (n = 5), whereas ratios of angular velocities were only of 1.05 and 1.06 for cilia close to captured particles at 26 and 27 [micro]m from the base of the cilium (n = 2).
Cilia were not detectably slowed when pushing a cell of I galbana. Angular velocities for the two cilia traveling with algal cells were 5.6[degrees] and 6.8[degrees] m[s.sup.-1] only slightly lower than the angular velocities of 6.2[degrees] to 7.8[degrees] m[s.sup.-1] for the five cilia approaching, leaving, or passing algal cells (mean [+ or -] SDof 7.1[degrees] [+ or -] 0.5[degrees] m[s.sup.-1]). Mean [+ or -] SD for angular velocities for 20 cilia with no particles nearby was 7.2[degrees] [+ or -] 0.5[degrees] m[s.sup.-1].
Speeds of particles increased with distance from the zone of recovery strokes to the tips of effective strokes, with a decrease at and beyond the distance reached by tips of cilia (Fig. 4). Estimates of particle speeds were similar for starting points at different parts of the effective stroke. Estimates grouped by portion of the arc are represented by different symbols in Figure 4A. The speeds were measured over 8-ms intervals so that the paths over which velocities were measured extended 16[degrees]--51 [degrees] beyond the starting point. At a given distance from the base of the cilia, the particles that were in contact with cilia at both beginning and end of the 8-ms interval were among the fastest (Fig. 4B). Particles in close contact with a cilium only at the beginning or end of the interval or that were passed by a cilium during the interval had a wide range of speeds. Particles that were not in contact with a cilium were among the slowest, especially those that passed distally (Fig. 4B). Proximity to a cilium, duration of proximity during an 8-ms interval, and velocities of the cilia all vary and affect velocities of particles. The mid-range of speeds for particles passing proximally, near the recovery strokes, was about 1.1 mm [ms.sup.-1] and for particles passing near the tips of cilia was about 3.5 mm [ms.sup.-1].
[FIGURE 4 OMITTED]
We compared velocities for the individual described above to velocities for three other larvae to check for individual differences. Estimates of ratios of velocities of cilium and particle for the individual analyzed in detail fell within the range of those for the other individuals (mean [+ or -] SD with n = 10: 1.5 [+ or -] 0.4 compared to 1.8 [+ or -] 0.5, 1.4 [+ or -] 0.4, and 1.2 [+ or -] 0.2). For all four individuals, particle velocities were consistently either slower than nearby cilia or matching the velocity of a cilium moving with the particle. Angular velocities of cilia were, however, greater for the individual that we analyzed in greater detail (mean [+ or -] SD with n = 10: 7.1 [+ or -] 1.0 compared to 4.3 [+ or -] 0.9, 3.8 [+ or -] 0.8, and 3.0 [+ or -] 0.4 in degrees per millisecond). This variation in velocities of cilia could reflect differing responses to tethering. The individual analyzed in detail appears to have been representative of the other tethered larvae, except for its faster effective strokes and its capture of more of the passing particles.
We obtained high-speed video recordings of the transport of more than 40 captured particles from the site of capture toward the mouth in the region between the prototroch and metatrochal ciliary bands, the food groove. A sequence of still images from high-speed video of such a transport event is shown in Figure 5. The sequence shows a particle (a polystyrene bead 5 [micro]m in diameter) captured on the left side of the larval body being rapidly transported toward the mouth and ingested at the left corner of the mouth. Similar capture and transport of an algal cell is shown in Figure 6, in ventral-posterior view. Figure 6 also shows that the spacing of prototrochal cilia in active stroke during particle captures was [approximately equal to]25 [micro]m.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Larval feeding performance
We estimated clearance rates of larvae of S. cementarium of three ages (and sizes: Table 1) on spherical particles ranging in diameter from 3 to 45 um (Fig. 7). We compared three aspects of feeding performance among these larvae--maximum clearance rates, particle size on which maximum clearance rates were obtained, and range of particle sizes that were ingested. Because clearance rate data did not meet assumptions of normality or equal variance even after transformation, we could not analyze them using two-way ANOVA as we had planned. We know of no nonparametric analog of a two-way ANOVA. Thus we do not present statistical tests of hypothesized differences in clearance rate among ages and bead sizes, and instead interpret the data graphically, as qualitative indicators of feeding performance. Comparisons of the clearance rates of 9-day-old and 30-day-old larvae suggest increases related to age (or size) in all three of these aspects of feeding performance; clearance rates of 21-day-old larvae were unexpectedly low and more difficult to interpret. Estimated ciliary band length-specific clearance rates are slight underestimates, as they were calculated by assuming that band length was equivalent to circumference at the prototroch, but a small portion of that circumference was interrupted by a dorsal gap in the feeding ciliation (Fig. 1C).
[FIGURE 7 OMITTED]
Table 1 Mean dimensions ([+ or -] standard deviation) of larvae of Sabellaria cementarium at three ages Age (d) Body Body Prototrochal length diameter cilia ([mu]m) ([mu]m) length ([mu]m) 10 122.6 [+ or -] 6.9, 112.3 [+ or -] 7.1, 44.6 [+ or -] 4.4, n = 7 n = 6 n = 8 21 162.8 [+ or -] 12.2, 140.9 [+ or -] 12.1, 47.0 [+ or -] 3.3, n = 9 n = 9 n = 9 31 278.4 [+ or -] 34.4, 166.4 [+ or -] 27.7, 55.2 [+ or -] 4.1, n = 6 n = 6 n = 6 Larvae were videotaped for measurement the day after clearance rates were estimated. Values for body length and diameter are means of n larvae; three estimates of prototrochal cilium length were taken per larva and averaged to provide a single value for that larva; the value shown is the mean of these estimates.
Nine-day-old larvae ingested particles ranging in diameter from 3 to 20 [micro]m and had their maximum clearance rates (0.77 m[l*d.sup.-1]) on particles 10 [micro]m in diameter. The mean length of the prototrochal ciliary band in these larvae (calculated from total body circumference without correction for the dorsal gap in ciliation) was [approximately equal to]353 [micro]m, leading to a band-length-specific maximum clearance rate of 0.0022 m[l*d.sup.-1] [micro] [m.sup.-1] (for beads of 10 [micro]m in diameter).
Clearance rates in 21-day-old larvae showed no clear pattern of particle size-specific maximum clearance rate; in these larvae, clearance rates were [approximately equal to]0.35 m[l*d.sup.-1] on particles 3-20 [micro]m in diameter. However, relative to younger larvae, 21-day-old larvae did demonstrate an increase in the maximum particle size that could be ingested--ingesting a few of the largest beads offered (45 [micro]m).
The oldest larvae examined, 30 days of age, ingested particles of all sizes up to the maximum offered (45 [micro]m). They fed at rates comparable to those of younger larvae on particles 3-10 [micro]m in diameter, but had much higher clearance rates on particles 15 and 20 [micro]m in diameter (1.16 and 1.37 m[l*d.sup.-1], respectively). Thirty-day-old larvae cleared 45-[micro]m beads at the rate of 0.21 ml*[d.sup.-1]. The mean length of the prototrochal ciliary band in these larvae (calculated as above) was [approximately equal to]523 [micro]m, leading to a band length-specific maximum clearance rate of 0.0026 m[l*d.sup.-1]* [mu] [m.sup.-1] (for beads of 20 [micro]m in diameter).
Previous authors have not identified metatroch and food groove ciliary bands, which are required for opposed-band feeding, in the larvae of sabellariids (e.g., Wilson, 1929; Dales, 1952; Cazaux, 1964; Smith and Chia, 1985). Some of these studies were quite detailed and extensive. It is thus not surprising that the absence of evidence of these ciliary bands in larvae of sabellariids has sometimes been interpreted as evidence of their absence (e.g., Rouse, 1999). However, the observations described here clearly demonstrate the presence of opposed bands of cilia in the feeding larvae of Sabellaria cementarium and Phragmatopoma cali-fornica. Further evidence indicates that opposed bands are also present in the larvae of S. alveolata. First, Wilson (1968) published a low-magnification micrograph of the larva of that species that shows likely metatrochal cilia posterior to the prototroch (his fig. 1A). Second, R. R. Strathmann (1987) also reported opposed-band feeding by sabellariid larvae (without presenting the supporting data), based in part on cinefilms of larvae of S. alveolata that showed capture of particles between prototrochal and metatrochal ciliary bands and their transportation to the mouth along a ciliated food groove. Thus several species of sabellariid annelids feed using opposed bands of cilia. Larvae of other species in the family remain to be studied with respect to feeding by ciliary bands.
The demonstration of opposed bands in larvae of sabel-lariids, as well as the results of several other recent studies in which opposed bands were described in otherwise well-known larvae (Pernet, 2003; Pernet and McArthur, 2006), suggests that studies of the morphology of annelid larvae should be interpreted very carefully when making inferences about the presence or absence of opposed bands. Many older studies of larval form, for example, though careful and detailed (e.g., Wilson, 1929, 1968), were carried out at a time when the functional significance of metatrochal and food groove cilia were not well known. Thus, their authors may simply not have paid attention to metatrochal cilia, or may have interpreted them as a posterior row of prototrochal cilia. Further, cryptic opposed bands (with short metatrochal cilia and narrow food grooves) may be present in nonfeeding larvae (e.g., Serpulidae: Nishi and Yamasu, 1992, their fig. 3C; Sabellidae: Rouse and Fitzhugh, 1994; Pernet, 2003); these are likely to be missed unless the observer is specifically looking for them.
Feeding kinematics and performance
Details of the kinematics and performance of larvae of S. cementarium are consistent with those reported from other opposed-band feeders. The observation that prototrochal cilia of S. cementarium traveled faster than particles and thereby overtook and traveled with them toward the food groove is similar to observations made in mollusc larvae that feed with opposed prototrochal and metatrochal bands (Strathmann and Leise, 1979; Gallager, 1988; Emlet, 1990; Romero et al., 2010). Romero et al (2010) demonstrated that particle capture can be by direct interception (Rubenstein and Koehl, 1977; LaBarbera, 1984; Shimeta and Jumars, 1991), with adhesion of a particle to a single cilium. This capture mechanism has also been discussed for opposed-band feeders as the "catch up principle" (Riisgard et al., 2000). Capture of particles passing through the prototrochal band close to the recovery strokes and retention of particles in the food groove at the end of prototrochal effective strokes may involve other processes as well (Gallager, 1988; Romero et al., 2010).
The capture of particles only within the proximal portion of the prototrochal cilia, as observed for S. cementarium, is consistent with some observations on mollusc veligers (Strathmann and Leise, 1979; Gallager, 1988), but captures within that zone may not represent the maximum capability of mollusc larvae or of larvae of S. cementarium. Veliger larvae of a gastropod captured particles almost to the tips of the prototrochal cilia (Romero et al., 2010). Observations of low capture rates and capture only in the proximal part of the area swept by cilia could indicate that the larvae have behaviorally reduced their clearance rate. One hypothesis is that opposed-band feeders can regulate rates of capture by changing the adhesiveness of the prototrochal cilia (Romero et al., 2010). Another possibility is that captures can be reduced by a reduction in the beat of the metatroch. Larvae of S. cementarium sometimes stopped beat of the metatroch while beat of the prototroch continued. Stopping of the metatroch while the prototroch continued was reported by Strathmann et al. (1972) for a serpulid larva, but was not subsequently observed in studies of mollusc larvae. Capture rates were low for the tethered larvae. If the larvae can capture all particles close to a cilium in its effective stroke, the maximum capture rate is much greater.
The larva observed with video (Fig. 3) and the 30-day-old larvae observed for ingestion rates (Fig. 7, Table 1) had similar cilium lengths and provide boundaries on estimates of maximum clearance rates. Clearance rates estimated from observations of particle ingestion are usually lower than the maximum rate possible for a larva. The same is true for a clearance rate estimated from the captures by the tethered larvae. An estimate can be obtained, however, from volume flow calculated from the particle speeds in Figure 4 under two assumptions: (1) that all particles will be contacted by a cilium while passing through the zone of prototrochal effective strokes; (2) that the particle will be captured if it contacts a prototrochal cilium anywhere from just beyond the zone of recovery strokes to near its tip. The speeds near the recovery strokes (at 17 [micro]m) and near the tips of cilia (at 51 [micro]m) were about 1.1 and 3.5 [micro]m m[s.sup.-1] (Fig. 4A, B). An estimate of volume flow is a speed of 2.3 [micro]m m[s.sup.-1] multiplied by the area of a ring with inner diameter of 200 [micro]m and outer diameter of 268 [micro]m. The ring corresponds to the diameter of the 30-day-old larva in Table 1 and distances of 17 and 51 [micro]m outward, along observed cilium lengths. That flow is 2.3 mm [s.sup.-1] multiplied by 0.025 [mm.sup.2] = 0.0575 m[m.sup.3] [s.sup.-1] = 4.97 ml [d.sup.-1]. (This is a slight overestimate, as it is calculated assuming that the entire circumference of the larva bears prototrochal cilia, ignoring the small dorsal gap: Fig 1C). The maximum clearance rate may fall between this estimate of about 5 ml [d.sup.-1] and the estimate of 1.37 ml [d.sup.-1] from ingestion rates of a 30-day larva.
The estimates of maximum clearance rates for larvae of 5. cementarium are similar to those for other small larvae that feed with opposed bands. For example, for 3-day-old trochophores of the capitellid Mediomastus fragile (presumed to feed with opposed bands), maximum clearance was [approximately equal to]0.23 ml [d.sup.-1] (Hansen, 1993). For veligers of the opistho-branch Philine aperta, with a shell length of about 270 [micro]m and a prototrochal cilium length of 50 to 55 [micro]m, maximum clearance rate was [approximately equal to]1.4 ml [d.sup.-1] (Hansen, 1991). These comparisons are limited by lack of comparable measures of size. Data that include estimated maximum clearance, length of the prototrochal ciliary band, prototrochal cilium length, and body weight are too scarce for interspecies comparisons of scaling of clearance rates by opposed bands.
One basis of comparison, however, is the velocities of particles through the prototrochal band near the tips of prototrochal cilia. If, at maximum clearance rates, particles are captured nearly to the tips of prototrochal cilia, then maximum clearance rates per unit band length depend on the cilium lengths and velocities. For larvae of S. cemen-tarium with prototrochal cilia about 50 to 55 [micro]m long, estimated particle velocities near the tips of prototrochal cilia were [approximately equal to]3.5 mm [s.sup.-1]. That estimate agrees with a comparison among species, in which particle velocities increased nearly in proportion to lengths of prototrochal cilia from about 30 to 100 [micro]m (Emlet and Strathmann, 1994). Insofar as maximum clearance rates depend on volume flow through the prototroch, the performance of larvae of S. cementarium is similar to that of other larvae that feed with opposed bands.
Distribution of opposed-band feeding in annelids
Our observations bring the total number of annelid families known to include larvae with opposed bands to 10 (Table 2). For three of these families--Amphinomidae, Capitellidae, and Polygordiidae--the published evidence is scant. Additional published evidence in the form of film or video footage of particle capture and transport will eventually resolve these cases. It is also possible that opposed bands of cilia will eventually be described in members of a few additional families of annelids. It is important to note that opposed bands of cilia may be present, and should be looked for, even in nonfeeding larvae. Pernet (2003), for example, described their presence in several species of Sabellidae, and Nishi and Yamasu (1992, their fig. 3C) demonstrated their presence in the nonfeeding larva of the serpulid Salmacina dysteri. The origin and functions, if any, of opposed bands in nonfeeding larvae remain to be understood.
Table 2 The ten families of annelids in which the larvae of one or more species are known to have opposed bands of cilia capable of capturing particles and transporting them to the mouth Annelid family References Amphinomidae * Feeding: Jagersten (1972): Kudenov(1974); see fig 12.13A.B in Pernet et al. (2002) Capitellidae * Feeding: Pernet & Schroeder 1999 [abstract only]; Pernet,unpubl. data; see fig 12.15B in Pernet et al. (2002) Echiuridae Feeding: Hatschek (1880); Miner et al (1999) Opheliidae Feeding: Miner et al (1999) Oweniidae Feeding: Emlet & Stralhmann (1994) Polygordiidae * Feeding: Hatschek (1878); Woltereck (1904); Pernet, pers. obs. Sabellidae All nonfeeding: Rouse & Fitzhugh (1994); Pernet (2003) Sabellariidae Feeding; this study Serpulidae Feeding; Strathmann et al. (1972);nonfeeding: Nishi & Yamasu (1992, fig. 3C) Spionidae Feeding; Pernet & McArthur (2006) In all cases but those of the sabellids (Pernet, 2003) and some serpulids (e.g., Nishi and Yamasu, 1992), the studied larvae are feeding. * Additional clear evidence (especially published film or video footage of particle capture or transport) is required for the three families identified with asterisks, but published descriptions and unpublished observations of living larvae are currently convincing to the present authors.
Accurate data on the distribution of opposed bands among the annelids are critical for inferences on how many times this larval feeding mechanism has been gained or lost in the clade (e.g., Rouse, 1999, 2000; Nielsen, 2001). The data we report here are a step toward this goal. However, other information required for such ancestral state reconstructions is still lacking. First, there is currently no strongly supported phylogenetic hypothesis describing relationships among the annelid families. Phylogenies based on morphological characters (e.g., Rouse and Fauchald, 1997) conflict with phylogenies based on molecular data (e.g., Struck et al, 2007; Zrzavy et al., 2009). Without a strongly supported family-level phylogeny, it is difficult to carry out reliable reconstructions of ancestral character state for the annelids as a whole.
Second, in at least four of the families listed in Table 2--Capitellidae, Sabellidae, Serpulidae, and Spionidae--there is within-family variation in the presence of opposed bands. For example, in the Spionidae, larvae of only one species (Streblospio benedicti: Pernet and McArthur, 2006) are currently known to possess opposed bands, while the feeding larvae of several other species clearly lack metatrochal cilia (e.g., Pernet et al, 2002). Further, for most of these taxa there are no strongly supported phylogenies describing relationships among genera and species. Together, this means that it is difficult to evaluate hypotheses on whether the common ancestor of each family-level clade possessed opposed bands. Without information on the ancestral character state for each family, it is difficult to carry out trustworthy analyses at higher levels.
However, substantial progress has recently been made in reconstructing relationships within some restricted clades of annelids. Of particular relevance to this study is the work of Kupriyanova and Rouse (2008), who presented a well-supported molecular phylogeny of some members of the traditional families Sabellidae and Serpulidae. Their phylogeny also included as terminal taxa members of the Oweniidae, Sabellariidae, and Spionidae, as other recent molecular phylogenetic analyses have suggested that these families are closely related to the Sabellidae and Serpulidae (Struck et al, 2007; Zrzavy et al, 2009). A simple parsimony analysis using Kupriyanova and Rouse's (2008) phylogeny (collapsing branches to the family level, but leaving the Fabriicidae and Sabellidae sensu stricto as separate clades) and data from Table 2, implemented in Mesquite 2.74 (Maddison and Maddison, 2010), suggests that opposed bands of cilia were present in the common ancestor of all the families listed above, but lost at least once, in members of the Fabriciidae. However, this analysis suffers from some of the problems noted above, and more robust analyses await the inclusion of additional taxa in the phylogenies, as well as additional data on the presence of opposed bands in the larvae of sampled taxa.
Even with improved taxon sampling and more robust phylogenetic hypotheses, however, inferences on ancestral characters also rely on models of evolutionary transitions. It is currently difficult to assess the degree to which the evolution of opposed-band feeding is facilitated by an ancestral prototroch that functioned only in swimming, by an ancestral prototroch that functioned in feeding without a metatroch, or by prior existence of an opposed-band feeding mechanism that was lost in an immediate ancestor. Also, the conditions under which a metatroch and food groove are retained after loss of the need to feed (as in sabellids and some serpulids) are also unknown. Developmental and functional studies may indicate conditions that promote gain and loss of the opposed-band mechanism. It is not yet known whether there is no bias in gain and loss of opposed-band feeding (as assumed in inferences that simply minimize the total number of evolutionary transitions) or, if a bias exists, the probability of loss versus gain of opposed-band feeding.
This research was supported by NSF grant OCE0623102 to RRS, and by the Friday Harbor Laboratories (FHL) of the University of Washington. We thank the staff of FHL for their help, including collection of the worms by the crew of the RV Centennial. Facilities for scanning electron microscopy were provided by FHL and by the Institute for Integrated Research on Environment and Society (IIRMES) at California State University Long Beach. We are grateful to R. Emlet and two anonymous reviewers for suggestions that improved an earlier version of this manuscript.
Cazaux, C. 1964. Developpement larvaire de Sabellaria alveolata. Bull. Inst. Oceanogr. Monaco 62: 1-15.
Dales, R. P. 1952. The development and structure of the anterior region of the body in the Sabellariidae, with special reference to Phragmatopoma californica. Q. J. Microsc. Sci. 93: 435-452.
Emlet, R. B. 1990. Flow fields around ciliated larvae: effects of natural and artificial tethers. Mar. Ecoi Prog. Ser. 63: 211-225.
Emlet, R. B., and R. R. Strathmann. 1994. Functional consequences of simple cilia in the mitraria of oweniids (an anomalous larva of an anomalous polychaete) and comparisons with other larvae. Pp. 143-157 in Reproduction and Development of Marine Invertebrates. W. H. Wilson. Jr., S. A. Strieker, and G. L. Shinn, eds. Johns Hopkins University Press, Baltimore.
Gallager, S. M. 1988. Visual observations of particle manipulation during feeding in larvae of a bivalve mollusc. Bull. Mar. Sci. 43: 344-365.
Giangrande, A. 1997. Polychaete reproductive patterns, life-cycles and life-histories: an overview. Oceanogr. Mar. Biol. Annu. Rev. 35: 323-386
Hadfield, M. G., M. F. Strathmann, and R. R. Strathmann. 1997. Ciliary currents of non-feeding veligers in putative basal clades of gastropods. Invertebr. Biol. 116: 313-321.
Hansen, B. 1991. Feeding behaviour in larvae of the opisthobranch Philine aperta. II. Food size spectra and particle selectivity in relation to larval behaviour and morphology of the velar structures. Mar. Biol. 111: 263-270.
Hansen, B. 1993. Aspects of feeding, growth and stage of development by trochophora larvae of the boreal polychaete Mediomastus fragile (Rasmussen) (Capitellidae). J. Exp. Mar. Biol. Ecoi 166: 273-288.
Hart, M. W. 1991. Particle captures and the method of suspension feeding by echinoderm larvae. Biol. Bull. 180: 12-27.
Hart, M. W., and R. R. Strathmann. 1995. Mechanisms and rates of suspension feeding. Pp. 193-221 in Ecology of Marine Larvae, L. McEdward, ed. CRC Press, Boca Raton, FL.
Hatschek, B. 1878. Studien uber Entwicklungsgeschichte der Anneliden. Arb. Zool. Inst. Univ. Wien 1: 277-404.
Hatschek, B. 1880. Uber Entwicklungsgeschichte von Echiurus. Arb. Zool. Inst. Univ. Wien 3: 45-78.
Hatschek, B. 1885. Entwicklung der Trochophora von Eupomatus uncinatus, Phillippi (Serpula uncinata), Arb. Zool Inst. Univ. Wien 6: 121-148.
Jagersten, G. 1972. Evolution of the Metazoan Life Cycle. Academic Press, New York.
Keever, C. C, and M. W. Hart. 2008. Something for nothing? Reconstruction of ancestral character stages in asterinid sea star development. Evoi Dev. 10: 62-73.
Kudenov, J. D. 1974. The reproductive biology of Eurythoe complanata (Polychaeta: Amphinomidae). Ph.D. thesis. University of Arizona.
Kupriyanova, E. K., and G. W. Rouse. 2008. Yet another example of paraphyly in Annelida: molecular evidence that Sabellidae contains Serpulidae. Mol Phylogenet. Evol. 46: 1174-1181.
LaBarbera, M. L. 1984. Feeding currents and particle capture mechanisms in suspension feeding animals. Am. Zool. 24: 71-84.
Lehrke, J., H. A. ten Hove, T. A. Macdonald, T. Bartolomaeus, and C. Bleidorn. 2007. Phylogenetic relationships of Serpulidae (Annelida: Polychaeta) based on 18S rDNA sequence data, and implications for opercular evolution. Org. Divers. Evol. 7: 195-206.
Maddison, W. P., and D. R. Maddison. 2010. Mesquite: a modular system for evolutionary analysis. [Online]. Available: http://mesquiteproject.org. [2010, 26 October].
McEdward, L. R., and R. R. Strathmann. 1987. The body plan of the cyphonautes larva of bryozoans prevents high clearance rates: comparison with the pluteus and a growth model. Biol. Bull. 172: 30-45.
Miner, B. G., E. Sanford, R. R. Strathmann, B. Pernet, and R. B. Emlet. 1999. Functional and evolutionary implications of opposed bands, big mouths, and extensive oral ciliation in larval opheliids and echiurids (Annelida). Biol. Bull. 197: 14-25.
Nielsen, C. 2001. Animal Evolution: Interrelationships of the Living Phyla. Oxford University Press, Oxford.
Nishi, E., and T. Yamasu. 1992. Brooding and development of a serpulid tube worm Salmacina dysteri (Sedentaria: Polychaeta). Bull. Coll Sci Univ. Ryukyus 54: 107-121.
Pernet, B. 2003. Persistent ancestral feeding structures in nonfeeding annelid larvae. Biol Bull. 205: 295-307.
Pernet, B., and L. McArthur. 2006. Feeding by larvae of two different developmental modes in the annelid Streblospio benedicti. Mar. Biol 149: 803-811.
Pernet B., and P. C. Schroeder. 1999. Opposed ciliary bands and larval feeding in a capitellid annelids. Am. Zool 39: 39A.
Pernet, B., P.-Y. Qian, G. W. Rouse, C. M. Young, and K. J. Eckelbarger. 2002. Phylum Annelida: Polychaeta. Pp. 209-243 in Atlas of Marine Invertebrate Larvae, C. M. Young, ed. Academic Press, London.
Phillips, N. E., and B. Pernet. 1996. Capture of large particles by suspension-feeding scaleworm larvae (Polychaeta: Polynoidae). Biol. Bull 191:199-208.
Riisgard, H. U., C. Nielsen, and P. S. Larsen. 2000. Downstream collecting in ciliary suspension feeders: the catch-up principle. Mar. Ecol Prog. Ser. 207: 33-51.
Romero, M. R., H. C. P. Kelstrup, and R. R. Strathmann. 2010. Capture of particles by direct interception by cilia during feeding of a gastropod veliger. Biol. Bull. 218: 145-159.
Rouse, G. W. 1999. Trochophore concepts: ciliary bands and the evolution of larvae in spiralian Metazoa. Biol. J. Linn. Soc. 66: 411-464.
Rouse, G. W. 2000. The epitome of hand waving? Larval feeding and hypotheses of metazoan phylogeny. Evol. Dev. 2: 222-233.
Rouse, G. W. 2006. Annelid larval morphology. Pp. 141-178 in Reproductive Biology and Phylogeny of Annelida, G. W. Rouse and F. Pleijel, eds. Science Publishers, Enfield, NH.
Rouse, G. W., and K. Fauchald. 1997. Cladistics and polychaetes. Zool. Scr. 26: 139-204.
Rouse, G. W., and K. Fitzhugh. 1994. Broadcasting fables: is external fertilization really primitive? Sex, size, and larvae in sabellid polychaetes. Zool. Scr. 23: 271-312.
Rubenstein, D. E., and M. A. R. Koehl. 1977. The mechanisms of filter feeding: some theoretical considerations. Am. Nat. 111: 981-994.
Shimeta, J., and P. A. Jumars. 1991. Physical mechanisms and rates of particle capture by suspension feeders. Oceanogr. Mar. Biol. Annu. Rev. 29: 191-257.
Smith, P. R., and F.-S. Chia. 1985. Larval development and metamorphosis of Sabellaria cementarium Moore, 1906 (Polychaeta: Sabellariidae). Can. J. Zool. 63: 1037-1049.
Stone, C. J. 1989. A comparison of algal diets for cirripeded nauplii. J. Exp. Mar. Biol. Ecol. 132: 17-40.
Strathmann, M. F. 1987. Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle. 670 pp.
Strathmann, R. R. 1978. The evolution and loss of feeding larval stages of marine invertebrates. Evolution 32: 894-906.
Strathmann, R. R. 1987. Larval feeding. Pp. 465-550 in Reproduction of Marine Invertebrates, Vol. 9, A. C. Giese, J. S. Pearse, and V. B. Pearse, eds. Blackwell, Pacific Grove, CA.
Strathmann, R. R., and D. Griinbaum. 2006. Good eaters, poor swimmers: compromises in larval form. Integr. Comp. Biol. 46: 312-322.
Strathmann, R. R., and E. Leise. 1979. On feeding mechanisms and clearance rates of molluscan veligers. Biol. Bull. 157: 524-535.
Strathmann, R. R., T. L. Jahn, and J. R, C. Fonseca. 1972. Suspension feeding by marine invertebrate larvae: clearance of particles by ciliated bands of a rotifer, pluteus, and trochophore. Biol. Bull. 142: 505-519.
Struck, T. H., N. Schult, T. Kusen, E. Hickman, C. Bleidorn, D. McHugh, and K. M. Halanych. 2007. Annelid phylogeny and the status of Sipuncula and Echiura. BMC Evol. Biol. 7: 57.
Vargas, C. A., P. H. Manriquez, and S. A. Navarette. 2006. Feeding by larvae of intertidal invertebrates: assessing their position in pelagic food webs. Ecology 87: 444-457.
Wilson, D. P. 1929. The larvae of the British sabellarians. J. Mar. Biol. Assoc. UK 16: 221-269.
Wilson, D. P. 1968. Some aspects of the development of eggs and larvae of Sabellaria alveolata. J. Mar. Biol. Assoc. UK 48:367-386.
Woltereck, R. 1904. Beitrage zur praktischen Analyse der Polygordius--Entwicklung, nach dem Nordsee and Mittelmeer Typus. Wilhelm Roux Arch. Entwicklungsmech. Org. 18: 377-403.
Zrzavy, J., P. Riha, L. Pialek, and J. Janouskovec. 2009. Phylogeny of Annelida (Lophotrochozoa): total-evidence analysis of morphology and six genes. BMC Evol Biol. 9: 189.
BRUNO PERNET (1), * AND RICHARD R. STRATHMANN (2)
(1) Department of Biological Sciences, California State University, Long Beach, 1250 Bellflower Blvd., Long Beach, California 90840; and (2) Department of Biology and Friday Harbor Laboratories, University of Washington, 620 University Rd., Friday Harbor, Washington 98250
Received 29 November 2010; accepted 17 May 2011.
* To whom correspondence should be addressed. E-mail: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Pernet, Bruno; Strathmann, Richard R.|
|Publication:||The Biological Bulletin|
|Date:||Jun 1, 2011|
|Previous Article:||Swimming of Xenopus laevis sperm exhibits multiple gears and its duration is extended by egg jelly constituents.|
|Next Article:||Populations of symbiodinium muscatinei show strong biogeographic structuring in the intertidal anemone Anthopleura elegantissima.|