Development to metamorphosis of the nemertean pilidium larva
© Maslakova; licensee BioMed Central Ltd. 2010
Received: 6 September 2010
Accepted: 2 December 2010
Published: 2 December 2010
Skip to main content
© Maslakova; licensee BioMed Central Ltd. 2010
Received: 6 September 2010
Accepted: 2 December 2010
Published: 2 December 2010
The nemertean pilidium is one of the most notable planktotrophic larval types among marine invertebrates. The juvenile forms inside the larva from a series of isolated rudiments, called the imaginal discs. The development culminates in catastrophic metamorphosis, in which the larval body is consumed by the juvenile worm. Although the pilidium was first described in 1847, and is commonly found among marine plankton, there is not a single complete description of its development. The few published studies of pilidial development are based on observations of typically unidentified larvae opportunistically collected from plankton at various developmental stages.
The development of Micrura alaskensis, a common Northwest Pacific coast intertidal nemertean, is described from fertilization to metamorphosis. A staging scheme is proposed based on characteristic developmental milestones. Three pairs of imaginal discs develop as invaginations of larval epidermis. The cephalic discs invaginate from the larval epidermis above the ciliated band, while the cerebral organ discs and the trunk discs invaginate below the ciliated band. All paired imaginal disc invaginations are closely associated with different portions of the larval ciliated band. In addition, two unpaired rudiments contribute to the juvenile - the proboscis rudiment and the dorsal rudiment, which do not develop as invaginations. A pair of thick-walled esophageal pouches previously thought to represent nephridial rudiments give rise to the juvenile foregut. Branched rudiments of protonephridia, and their efferent ducts are also described. Larval and juvenile serotonergic nervous systems are briefly described. Development of the juvenile is completed by 5-8 weeks at 11-15 degrees C. During the rapid metamorphosis the juvenile emerges from and devours the larva.
This study is the first description of pilidial development from fertilization to metamorphosis in a single species. It is illustrated with photomicrographs of live larvae, diagrams, confocal images, and videos. The findings are discussed in the context of previously published accounts of pilidial development, with which they disagree on several accounts. The results described here indicate a different number, origin and fate of various juvenile rudiments. The proposed staging scheme will be useful in subsequent studies of pilidial development.
The ciliated planktotrophic pilidium larva of nemertean worms (phylum Nemertea) is one of the most characteristic and unmistakable representatives of marine plankton. It is shaped like a helmet and equipped with an apical tuft of long cilia, and, typically, four lobes, spanned by the larval ciliated band . The thin-walled and funnel-like pilidial esophagus (technically a vestibule) is separated from the thick-walled globose stomach by a sphincter (which could be considered as the pilidial mouth). The larval gut ends blindly. The larva swims with the apical organ pointing forward, and the larval antero-posterior axis corresponds to the animal-vegetal axis of the zygote. The pilidium larva, first described by Müller in 1847 , looks nothing like the juvenile nemertean worm, which Metschnikoff discovered in 1869  develops inside the larva from a series of isolated rudiments called the imaginal discs. In a typical pilidium the antero-posterior axis of the juvenile is at a 90° angle with respect to the antero-posterior axis of the larva. The unusual development of the pilidium culminates in a catastrophic metamorphosis during which the juvenile worm emerges from and devours the larval body [4, 5]. Although the term "metamorphosis" has been applied, particularly in the older literature, to the entire process of development of the juvenile worm inside the pilidium larva, I am using it in a more restrictive sense, referring to the end-point of pilidial development - i.e. the rapid transition between the pilidium and the juvenile.
The pilidium larva is found in a single clade called the Pilidiophora , which comprises roughly a third of all nemertean species , and represents a derived mode of development for the phylum . Other nemerteans possess uniformly ciliated planula-like planktotrophic or lecithotrophic larvae, or direct development [1, 9, 10]. Pilidia belonging to different species vary in shape, size and color, but few have been matched to their respective adults [5, 10–12]. Modifications of pilidial development include Desor's larva in Lineus viridis and Schmidt's larva in Lineus ruber, Iwata's larva in Micrura akkeshiensis and a few recently discovered others [[10, 16], Megan Schwartz (University of Puget Sound) pers. comm.]. Development of Desor's and Schmidt's larva is encapsulated, while the others have planktonic lecithotrophic development. In all of these cases, the pilidial lobes are lacking, the blastocoel is reduced, the larva does not feed, but, similar to the canonical pilidium the juvenile develops via imaginal discs.
Although the development of the juvenile from imaginal discs is surely the most remarkable feature of the pilidium, this process remains poorly understood. The larval epidermis is very thin and transparent, making it easy to observe juvenile rudiments in live larvae. However, because the larvae are planktotrophic and the development of the juvenile takes weeks to months, the few studies of pilidial development are limited to descriptions of early stages, i.e. before onset of rudiment formation [12, 17–20], or observations based on unidentified pilidium larvae opportunistically collected from plankton [21–23]. The only published report of pilidium larvae raised from fertilization to the point of formation of a nearly complete set of imaginal discs is that by Schmidt  for Cerebratulus marginatus (see  for German translation). Most of what we know about the development of the juvenile inside the pilidium comes from the 1912 study by Salensky .
On the other hand, the development of the juvenile inside the encapsulated Desor's "larva" has been studied in detail by many authors [13, 26–29]. These embryos are opaque, so the observations are based on studies of preserved and sectioned material. Because this mode of development is derived for Pilidiophorans, it is not clear how much of what is known about the juvenile development inside Desor's larva is peculiar to it, and how much pertains to the pilidial development. For example, Desor's larva is described to have a total of eight imaginal discs, including a separate proboscis rudiment, whereas the typical pilidium is considered to have a total of seven discs, with proboscis developing as a derivative of the cephalic discs [ and references therein, and modern textbooks, e.g. [1, 9, 30]], although see Bürger  and Schmidt [24, 25].
This paper describes the development of a typical pilidium larva in M. alaskensis from fertilization to metamorphosis, focussing particularly on the relative timing of developmental events, and the number and origin of the various juvenile rudiments. A staging scheme is proposed based on characteristic developmental milestones. Larval and juvenile serotonergic nervous systems are briefly described. The results are discussed in the light of previously published accounts of pilidial development.
Using the method described here, I successfully reared to metamorphosis dozens of cohorts of pilidium larvae of Micrura alaskensis.
The rate of development depends on the temperature, feeding regime, and, possibly, other factors. Development of M. alaskensis in culture is somewhat asynchronous, even among full siblings raised in the same container. However, the juvenile rudiments always appear in the same sequence: first a pair of cephalic discs (Figure 3B), then a pair of trunk discs (Figure 3C), followed by the unpaired proboscis rudiment (Figure 3D), a pair of cerebral organ discs (Figure 3E), and the unpaired dorsal rudiment (Figure 3F). I observed two pairs of imaginal discs, the cephalic and the trunk discs, as early as 9 days after fertilization, although, more typically, at 14-15 days. The proboscis rudiment appears at about the same time as the cerebral organ discs, as early as 14 days of development, but more typically around 20 days.
Additional file 1: Movie 1: Metamorphosis of the pilidium larva of Micrura alaskensis. This time-lapse movie shows a typical sequence of pilidial metamorphosis, in which the juvenile successively breaks through the amniotic membrane and the larval epidermis, then backs out of the larval body while simultaneously consuming the larval tissues. In this sequence, the entire larval body is devoured within a few minutes, and the young nemertean worm crawls away having finished its first meal. The video is sped up 2.5× compared to real time. (MOV 20 MB)
Description of larval stages of the pilidium of Micrura alaskensis based on certain developmental milestones.
Earliest age (PF) observed in culture
flat square blastula, swims with cilia (Figure 2A-B, D)
radially symmetrical swimming stage with an apical tuft and a vegetal invagination (Figure 2C)
bilateral symmetry apparent in the gut, longer cilia along the vegetal margin, nascent lateral lappets (Figure 2E-H)
helmet-shaped pilidium with a ciliated band spanning the lappets, capable of capturing unicellular algae (Figure 3A)
cerebral organ discs
head and trunk
the proboscis rudiment fused with the cephalic discs to form the head rudiment; cerebral organ discs fuse with the respective trunk discs to form the trunk rudiment; an unpaired dorsal rudiment appears near the larval stomach, it is initially separate from the trunk discs and later fuses with them (Figure 3F-G, 9E)
dorsal margin of the juvenile trunk extended over the posterior portion of proboscis, but has not yet reached and fused with the dorsal margin of juvenile head (Figure 4B)
Once the invaginated rudiments of cephalic and trunk discs separate from the larval epidermis, they consist of two layers separated by a narrow cavity, like a flattened balloon. The thick inner layer gives rise to the juvenile tissues, including the definitive epidermis, and the thin outer layer composed of a single layer of squamous cells gives rise to the amniotic sack, or amnion. The amnion produced by the fused thin outer walls of cephalic and trunk imaginal discs encloses the juvenile inside the pilidium larva and separates it from the blastocoel. In Micrura alaskensis the amnion is characteristically decorated with black and brown pigment granules (Figures 3E, G-I, 4A, C, F-H, 5), which become apparent even before the imaginal discs fuse, and are evident in metamorphosing larvae. The black pigment granules are concentrated along the ventral side of the amnion, while the brown pigment granules form a polka-dot pattern on the lateral sides.
The dorsal rudiment, like the proboscis rudiment, is first evident as a cluster of what seems to be few mesenchymal cells in the plane of bilateral symmetry, at the diametrically opposite end from the proboscis. It is sandwiched between the epidermis of the posterior lobe and the stomach (Figure 3F). It subsequently fuses with the trunk discs to form the trunk rudiment (Figure 3G).
The juvenile inside the pilidium possesses a separate serotonergic nervous system, which becomes evident after torus stage. As in adult nemerteans, the most prominent component of the juvenile nervous system includes the two lateral nerve cords that originate from the cerebral ganglia and are connected anteriorly via dorsal and ventral brain commissures and posteriorly in the juvenile caudal cirrus (Figure 11). In addition, one can distinguish the juvenile subepidermal nerve network, and a number of longitudinal nerves in the proboscis (Figure 11). The only component of the larval serotonergic nervous system that appears to be incorporated into the juvenile is the oral nerve.
The results of this study confirm classical reports that the typical pilidium larva possesses three pairs of imaginal discs which develop as ectodermal invaginations -- the cephalic discs, the trunk discs, and the cerebral organ discs. This study disagrees with the only other detailed report on the order of appearance and the origin of these discs in pilidium larvae , and confirms Schmidt's report for Cerebratulus marginatus[24, 25]. Salensky, who relied solely on planktonic samples reported that all three pairs of imaginal discs appear simultaneously [: p. 21]. Schmidt [24, 25] reared pilidia of C. marginatus from fertilization to the point of formation of almost all of the juvenile rudiments and reported that the discs appear in the following sequence: cephalic discs first, followed by the trunk discs, and finally, the cerebral organ discs. The present study shows that in M. alaskensis, the appearance of the paired imaginal discs follows the same order: the cephalic discs appear after about one week of development, followed by the trunk discs at two weeks, and the cerebral organ discs at about three weeks of development at ambient sea temperature.
My unpublished observations of development in another Northwest Pacific coast species Cerebratulus cf. marginatus (which is likely not the same species as Schmidt's C. marginatus from the Gulf of Naples, Italy), as well as in numerous pilidia belonging to different species captured in plankton in coastal waters of Washington and Oregon suggest that they all follow the same sequence of appearance of imaginal discs. Although the development of the juvenile was not the focus of his study, Cantell's observations of pilidium larvae of Lineus albocinctus, Lineus bilineatus, Micrura purpurea, and several unidentified pilidial morphotypes collected from plankton  suggest the same order of imaginal disc formation as described by Schmidt [24, 25] and in this study. It is easy to see, however, that if one is relying solely on the larvae obtained from plankton, as opposed to rearing them in the lab, one might be misled into thinking that all three pairs of imaginal discs appear simultaneously, if intermediate stages of development did not present themselves to the collector.
According to textbooks, the juvenile inside the pilidium larva develops from seven imaginal discs: the paired cephalic discs, paired trunk discs, paired cerebral organ discs, and an unpaired dorsal disc [1, 9, 30]. This is largely based on the most detailed published study of development of the juvenile inside the pilidium larva by Salensky . Eight imaginal discs, including a separate proboscis rudiment are described in development of Desor's larva [26–29]. According to Salensky , the proboscis of the pilidium larva develops from the fused cephalic discs, and a separate proboscis rudiment is lacking. Interestingly, Bürger  and Schmidt [24, 25] described a separate proboscis rudiment in development of the pilidium larva, but this observation was not confirmed by Salensky  and is ignored in the more recent literature.
The results of the present study clearly show that in the pilidium larva of Micrura alaskensis there are eight separate juvenile rudiments, including an unpaired proboscis rudiment, which confirms earlier reports by Bürger for an unidentified species  and Schmidt for Cerebratulus marginatus[24, 25]. Because the proboscis rudiment is relatively small and is evident as a separate entity from the cephalic discs only for a short period of time, it is possible that Salensky and other investigators relying as they did on planktonic samples instead of culturing larvae, did not have the opportunity to observe and confirm its presence.
The proboscis rudiment in M. alaskensis does not develop as an invagination, contrary to Bürger's report . It is not clear whether it arises by delamination from the pilidial epidermis, or from a cluster of mesodermal cells associated with the pilidial epidermis. In the first case the proboscis rudiment would be derived from the first-quartet micromeres. In the latter it would most likely originate from the 4d cell, the spiralian mesentoblast. The 4d cell gives rise to a population of scattered mesenchyme cells in the pilidium of Cerebratulus lacteus, including a small cluster of cells in vicinity of the cephalic discs, i.e. where proboscis rudiment would be developing later (Figure 7d in ). It appears very likely that the bilayered proboscis bud described here (Figure 6E) has a dual origin, so that its inner part ("the arm") is derived from the fused cephalic discs, as proposed by Salensky , while the outer part ("the sleeve") from the separate proboscis rudiment described here. Conceivably, the separate rudiment is derived from the 4d cell and gives rise to the muscle layers of the proboscis and the rhynchocoel, while the portion derived from the cephalic discs gives rise to the glandular epidermis of the proboscis. This would make sense in the context of a now widely accepted hypothesis that the nemertean rhynchocoel is homologous to the annelid and mollusk coeloms , which are also derived from the 4d cell. A study utilizing long-lasting cell lineage markers is necessary to determine which is the case.
Another piece of evidence that the proboscis is not derived solely from the cephalic discs, as suggested by Salensky , comes from an incidental observation of development of one particular cohort of pilidium larvae reared in the lab, which exhibited numerous abnormalities in juvenile, but not larval, development [Maslakova unpublished]. In many of these pilidia one or several juvenile rudiments were missing, while others appeared to be unaffected. Some larvae had all of the discs, including both of the cephalic discs, but were missing the proboscis. This suggests that even if cephalic discs normally contribute to the proboscis, it is possible that they may need some sort of inductive signal from the unpaired proboscis rudiment in order to do so. It would be possible to test whether the anterior unpaired rudiment observed in this study is essential to the formation of the proboscis by experimentally destroying it (e.g. by laser ablation) in otherwise normally developing pilidium larvae.
Although textbooks often describe all of the juvenile rudiments as invaginations of larval epidermis, the results of this study suggest that only the paired imaginal discs develop as distinct invaginations, whereas the proboscis rudiment and the dorsal rudiment do not. Salensky  reported that the unpaired dorsal disc develops via delamination from the pilidial dorsal epidermis. The present study is inconclusive as to whether the dorsal disc is of mesodermal origin (i.e. likely derived from the 4d cell) or is in fact an epidermal derivative (i.e. likely derived from one of the first-quartet micromeres).
Several earlier studies describe a pair of esophageal pouches during juvenile development inside Desor's larva  and pilidium larvae [22, 23] as rudiments of the nephridia (although Nusbaum and Oxner  disagree with this interpretation). Bürger  reported that the two pouches completely separate from the esophageal cavity of the pilidium larva, fuse with the trunk discs, then branch like fingers of a glove, and that the nephridiopores develop after metamorphosis. Salensky  reported two pouches in a similar position (between the cerebral organ discs and the larval stomach), but interpreted them as being derived from the subumbrellar epidermis near esophagus, rather than the esophageal wall. He insisted on this particular distinction, because he believed that the pilidial esophagus is of endodermal origin, and the nephridia must be derived from the ectoderm. Histologically, however, there is no difference between the subumbrellar epidermis and esophageal epidermis; one gradually transitions into the other. Both are composed of squamous cells, and both are likely ingested by the juvenile during pilidial metamorphosis. Moreover, cell lineage analysis in the pilidium larva of Cerebratulus lacteus shows that the larval esophagus is derived from the second-quartet and third-quartet micromeres, i.e. is of ectodermal origin. More importantly, however, Salensky  did not observe these pouches closing off from the esophagus, even in the latest stages of juvenile formation inside the pilidium larva, nor did he observe any branching.
At the same time, I have also observed what appears to be a pair of protonephridia -- one on each side of the developing juvenile, in the immediate vicinity of the paired esophageal pouches described above, but not physically connected to them. These organs, provisionally referred to as the nephridia, are sandwiched between the cerebral organs and the stomach on two sides, and the esophageal pouches and the juvenile body wall on two other sides. Each nephridium has a branched distal portion, and at least one efferent duct leading through the body wall to the nephridiopore on the lateral side of the developing juvenile in advanced stages of M. alaskensis (Figure 12, Additional file 7). The position of these rudiments corresponds to that of protonephridia in adult nemerteans (restricted to the foregut region in most species). Because the nephridial rudiments are located in close proximity to the foregut rudiments, and because Bürger's study  was based on regular histological sections which are typically 7-8 μm thick (compared to 1-μm-thick confocal sections in this study), he might have been mislead into thinking that a) the nephridial rudiments are connected to the esophageal pouches, b) these pouches are closed off from the esophagus, and c) the nephridial openings develop only after metamorphosis. Future studies utilizing TEM of ultra-thin sections of advanced developmental stages of pilidium larvae should be able to confirm the nature of these provisional nephridia.
The results of this study largely confirm previous reports on the structure of the pilidial larval and juvenile nervous systems based on classical histological methods [22, 23], TEM  and immunohistochemistry and fluorescent microscopy .
Salensky  believed that the central nervous system of the juvenile (cerebral ganglia and lateral nerve cords) is derived from the cephalic imaginal discs, that there are no rudiments of the nervous system in the trunk discs, and that the lateral nerve cords invade the tissue of the trunk discs after fusion with the cephalic discs. Bürger  believed that the central nervous system is derived from both the cephalic imaginal discs and the trunk discs and that the juvenile nervous system develops at about the time when the cephalic and the trunk discs fuse with each other. According to Bürger  the cephalic discs give rise to the dorsal cerebral ganglia, while the trunk discs give rise to the ventral cerebral ganglia and the lateral nerve cords. The results of this study cannot confirm or disconfirm either hypothesis. It does appear very likely that at least some portion of the cerebral ganglia originates from the cephalic discs. However, a long-term lineage tracing is necessary to determine with confidence which imaginal discs give rise to various parts of the juvenile nervous system.
This study represents the first report documenting pilidial larval development from fertilization to metamorphosis. It is based on observations of laboratory cultures of the common Northwest Pacific coast species Micrura alaskensis. Larvae typically reach metamorphosis after 5-8 weeks of development at ambient sea temperature. During metamorphosis the juvenile worm escapes from the larval enclosure and routinely devours the entire pilidial body. The larval and juvenile serotonergic nervous system is for the first time illustrated using confocal microscopy.
Larval development is asynchronous in culture, so a staging scheme is proposed based on certain developmental milestones to facilitate comparison with other species. Characteristic developmental stages include, in that order: blastosquare, gastrula, young pilidium, feeding pilidium, cephalic discs, trunk discs, cerebral organ discs, head and trunk, torus, extended proboscis, complete proboscis, hood, metamorphosis. The order of appearance of juvenile rudiments is always the same and appears to be conserved between different pilidiophoran species.
The paired cephalic discs, paired trunk discs, and paired cerebral organ discs develop as invaginations of the larval epidermis. The cephalic discs invaginate from the larval episphere, while the trunk discs and the cerebral organ discs invaginate from the hyposphere. A separate unpaired dorsal rudiment previously described for pilidium larvae is also found in M. alaskensis, and does not develop as an invagination. The rudiments of juvenile nephridia and foregut, which have been described but confused with each other in the earlier literature and ignored by the modern literature are described and illustrated using confocal microscopy. One of the most significant findings of this study is the separate proboscis rudiment, which brings the total number of juvenile rudiments inside the pilidium larva from seven, as described in textbooks, to eight. The unpaired proboscis rudiment does not develop as an invagination, and may be of mesodermal, rather than ectodermal origin.
Reproductive adults of M. alaskensis were collected during negative low tides in False Bay (San Juan Island, Washington, USA) in July-August 2005, June-August 2006 and July 2007, and from several different mudflats in vicinity of Charleston, Oregon during the summer months of 2008-2010. Occasionally, ripe individuals were encountered outside the usual summer reproductive period, e.g. in February 2007 in False Bay and in March 2009 and March 2010 in Charleston, OR. Ripe males and females can be distinguished under the dissecting microscope, as testis and ovaries are visible through the body wall. Gametes were obtained by dissection. Several hundred to thousands of eggs can be obtained from a single female. Primary oocytes with distinct germinal vesicles dissected from the females or spawned naturally, undergo germinal vesicle breakdown soon upon contact with sea water, and can be fertilized by a dilute suspension of sperm obtained by dissection.
Developing embryos were cultured in 150 ml glass custard dishes in 0.45 μm-filtered seawater (FSW) until they started swimming. Swimming stages were transferred to larger volumes of FSW and cultured with continuous stirring using plexiglass paddles. Initial concentrations were approximately 1 larva per ml, and they were progressively thinned to about 1 larva per 4-10 ml over the course of several weeks of development. Water was changed by reverse filtration every three to four days, and after each water change larvae were fed either Rhodomonas lens at a concentration of 104 cells/ml, or a mixed diet of 5 × 103 cells/ml each of Rhodomonas lens and Isochrysis galbana.
Live pilidium larvae were gently trapped between the slide and the coverglass supported by clay feet, and photographed using Nikon Coolpix 4500 digital camera mounted on a Nikon Eclipse 600 equipped with DIC optics or a Leica DCF 400 digital camera mounted on an Olympus BX51 microscope equipped with DIC. The timelapse video of metamorphosis was captured with side illumination through the Nikon Eclipse 600 microscope equipped with a Sony 3CCD video camera.
Larvae were relaxed in a 1:1 mixture of 0.33 M MgCl2 and filtered seawater for 10-15 minutes at room temperature (RT) and fixed for 30 min at RT in 4% paraformaldehyde freshly prepared from 16% ultrapure paraformaldehyde (Electron Microscopy Sciences) and filtered seawater. Following fixation, larvae were either rinsed in three 10 min changes of phosphate buffer saline (PBS) and stored in PBS at 4°C, or immediately permeabilized and stained. For confocal microscopy larvae were permeabilized in three 10 min changes of phosphate buffered saline (PBS) with 0.1% Triton X-100 (PBT), stained with Bodipy FL Phallacidin (1 U per 100 μl of PBT) and a nuclear dye Hoechst 33342 (1 μM) for 40 min at RT, rinsed in three 10 min changes of PBS and mounted in Vectashield (Vector Laboratories) on Poly-L-lysine coated coverslips supported by foil tape over microscope slides. These semi-permanent preps were sealed with nail polish and viewed immediately or stored at 4 degrees C.
To visualize the serotonergic nervous system, larvae were fixed and permeabilized as described above, then blocked in 5% Normal Donkey Serum (Jackson Immunoresearch) in PBT for 2 hours at RT. After three 10 min washes in PBT, larvae were incubated for 2 hours at RT or overnight at 4 degrees C; with the rabbit-anti-5HT primary antibody (ImmunoStar Cat.# 20080) diluted 1:500 in PBT, washed in three 10 min changes of PBT and incubated for 2 hours at RT with Alexa Fluor 488 donkey-anti-rabbit secondary antibody (Molecular Probes) diluted 1:600 in PBT, adding 10 μg/ml of propidium iodide for the last 40 minutes of incubation to stain the nuclei. Stained larvae were washed in three 10 min changes of PBS and mounted as described above.
Stained samples were examined with a BioRad Radiance 2000 laser scanning confocal mounted on a Nikon Eclipse E800 microscope with a 40× 1.3 N.A. oil lens or an Olympus Fluoview 1000 laser scanning confocal mounted on an Olympus IX81 inverted microscope with a UPlanFLN 40× 1.3 NA oil lens or a PlanApoN 60× N.A. 1.42 oil lens. Stacks of 0.5-1 μm optical sections were imported into ImageJ v. 1.42i (Wayne Rasband, National Institutes of Health, Bethesda, MD, USA) for image processing.
A large part of this study was carried out at the University of Washington's Friday Harbor Laboratories, while I was supported by the FHL Postdoctoral Fellowship. I thank the FHL researchers and staff, and particularly Dr. Richard Strathmann, for pointing me to M. alaskensis, as a suitable species for long-term laboratory culture, and Dr. George von Dassow for help with videomicroscopy and image processing and comments on the manuscript. Ms. Laurel Hiebert assisted in rearing pilidium larvae at the Oregon Institute of Marine Biology (University of Oregon).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (<url>http://creativecommons.org/licenses/by/2.0</url>), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.