Reference: Biol. BUN. 185: 77-85. (August, 1993)
Larval Development (with Observations on Spawning)
of the Pencil Urchin Phyllacanthus imperialis: a New
Intermediate Larval Form?
RICHARD RANDOLPH OLSON’~2~3,J. LANE CAMERON2%4, AND CRAIG M. YOUNG2
‘Australian Institute of Marine Science, P.M.B. 3, Townsville, Queensland 4810, Australia, *Harbor
Branch Oceanographic Institution, 5600 Old Dixie Hwy., Ft. Pierce, FL 33450, 3Department of
Zoology, University of New Hampshire, Durham, NH 03801, and 4DAMES & MOORE, 500 Market
Place Tower, 2025 First Avenue, Seattle, WA 98121.
Abstract. Information for understanding the evolution- occur during the shift from planktotrophy to lecithotrophy
ary shift from feeding to nonfeeding in echinoderm larvae in echinoid larvae.
can be gained from species whose larval development
pattern appears to be intermediate between these ex- Introduction
tremes. In this paper we report the development of one The causesand consequencesof the evolutionary shift
such species. from feeding to nonfeeding larvae has become a popular
The pencil urchin Phyllacanthus imperialis spawned topic in recent years (Raff, 1987; Parks et al., 1988; Wray
synchronously with the mass spawning of scleractinian and Raff, 1991; Amemiya and Emlet, 1992; McMillan et
corals at Lizard Island, Australia, in two consecutive years. al., 1992). Among echinoids, feeding larvae (plankto-
Their large yolky eggs(507 pm diameter) developed into trophs) are considered to be the ancestral form, and non-
nonfeeding echinopluteus larvae with two pairs of larval feeding larvae (lecithotrophs) are thought to be more de-
arms. The arms were identified as postoral and postero- rived (Strathmann, 1974). Within this group of echino-
dorsal, which are the first and third pairs in typical echi- derms, the shift from planktotrophy to lecithotrophy is
noplutei. A larval skeleton was present, with skeletal rods believed to have evolved independently as many as 14
extending the length of the arms. Five primary podia of times (Emlet, 1990; <f Strathmann, 1974). Although the
the juvenile rudiment appeared at 2 days of age. Meta- potential costs and benefits of direct development have
morphosis of the larvae and settlement began 4 days after been well discussed(Vance, 1973; Christiansen and Fen-
fertilization. Histological examination of 2-day-old larvae chel, 1979; Obrebski, 1979) the reasons for shifts in de-
revealed the presence of a developing gut, but no mouth velopmental mode are not completely understood.
opened in what would be the oral region of a typical Planktotrophy among echinoid larvae is known to be
echinopluteus, or the oral surface of the juvenile rudiment correlated with egg size (Emlet et al., 1987). Of the 276
in older larvae. Like other cidaroid larvae, this species echinoids reviewed by Emlet (1990) 66% have plankto-
showed no evidence of an amniotic invagination. trophic development and develop from small eggs(~350
The larva of P. imperialis appears to be a transitional pm), whereas 20% have lecithotrophic development and
form between the morphology of feeding and nonfeeding develop from larger eggs(>500 pm).
echinoid larvae. Traces of the ciliary band in the oral re- Before our study, only three echinoids had been re-
gion and the presenceof arms typical of the echinopluteus ported with planktonic larvae that do not fall into one of
larva indicate its evolutionary past, whereas the large egg these two groups; Brisaster latifrons, which has a feeding
size and absenceof a mouth hint at its future. This larval echinopluteus that develops from an egg 345 pm in di-
form provides insights into developmental changes that ameter (Strathmann, 1979); Clypeaster rosaceus (egg di-
ameter = 280 pm), whose larvae are facultative plank-
Received 11 January 1993; accepted 17 May 1993. totrophs, having the echinopluteus form (Emlet, 1986);
78 R. R. OLSON ET AL.
and Peronella japonica (egg diameter = 276 pm), which als (cf: Babcock et al., 1986). Numerous P. imperialis were
produces a lecithotrophic “pluteus” with only two arms, observed “perched” on the tops of coral heads or mounds
but shows normal development of the echinus rudiment of rubble; one male P. imperialis was observed spawning.
(Okazaki and Dan, 1954). Here we report a fourth “in- In the days following collection, spawning was induced
termediate” larval form. in these urchins by injecting them with 0.55 M KCl.
The pencil urchin Phyllacanthus imperialis is a member Freshly spawned eggs were yellow-tan and embedded in
of the order Cidaroida, which is believed to be the most a clear gelatinous outer coating that was viscous and sticky.
primitive extant echinoid order (Paul and Smith, 1984). Spawned eggs were held together by this coating to form
It is common on coral reefs throughout the Indo-west a strand of eggs that floated directly to the surface as it
Pacific but, being nocturnal and cryptic, is rarely seen was released. Mean egg diameter was 507 pm (SD = 3 1.9,
even at night. Mortensen (1938), who collected P. imper- n = 10).
ialis in the Red Sea, reported its egg to be 500 pm in Additional urchins were collected on 9 December (five),
diameter, but was unsuccessful in attempting in vitro fer- and 13 December (three). None of these urchins spawned
tilization. In this paper we provide the first description of when injected with KCl; examination of their excised go-
the larva and development of P. imperialis, which has the nads revealed that all were spent.
echinopluteus form but is lecithotrophic. The following year ( 1987) a few adult P. imperialis were
observed early in November shortly after the full moon.
Materials and Methods The density of urchins was considerably lower, and the
urchins were not as exposed as had been noted the pre-
Gametes of P. imperialis were obtained from adults vious year. Approximately 10 urchins were collected over
collected at Lizard Island, Australia, during the summers several nights, but only males produced viable gametes
of 1986 and 1987. Urchins were spawned in the days fol- upon injection with KCl. The few females that did respond
lowing collection by injecting 5 to 10 ml of 0.55 A4 KC1 to KC1 injection released masses of what appeared to be
into the coelomic cavity. Eggs were fertilized within 1 h undifferentiated yolk material. Urchins collected in late
of spawning. Embryos and larvae were maintained in an November also were not ripe. No male P. imperialis
open 3-l beaker at room temperature (29-3 1“C) without collected at this time spawned when injected, and those
stirring. For histological and morphological examination, females that did spawn produced egg-like irregularly
embryos and larvae were preserved in Bouin’s solution at shaped masses of yolk. Urchins collected on 2, 3, 4, and
20 h and 2, 3, and 4 days after fertilization. To preserve 8 December and injected with KC1 all responded similarly.
the larval skeleton, a few larvae were preserved in buffered Full moon occurred on 5 December in 1987. On the
10% formalin. Embryos and larvae were prepared for morning of 9 December, two females injected with KC1
scanning electron microscopy (SEM) by dehydration in released large numbers of eggs, the majority of which were
a graded series of ethyl alcohol through amyl acetate, dried spherical and uniform in size. About a third of these eggs
at the critical point (CO,), gold coated, then examined were successfully fertilized. The following day ( 10 De-
and photographed using a Novascan 30 scanning electron cember) two remaining females were spawned, and vir-
microscope. Larvae for histological examination were de- tually all of their eggs were uniform in size and shape.
hydrated to 50% ethyl alcohol, then embedded in JB-4 Fertilization of these eggs was nearly 100%.
embedding medium (Polysciences, Inc.). Sequential sec-
tions 2-3 pm thick were cut with glass knives and stained
with Richardson’s stain. Larval skeletal pieces were ex- Development
posed by corrosion of the soft tissues with 5% sodium Oocytes were opaque (yellow-tan in color) and floated
hypochlorite (household bleach) from 4-day larvae fixed at the surface of the water in their culture beakers. A dis-
in buffered formalin. tinct fertilization membrane was noted 15 min after the
introduction of sperm. First cleavage was observed 15 min
Results later-i.e., 30 min after fertilization (Table I). Ten hours
after fertilization the embryos were ciliated blastulae ro-
tating within the fertilization membrane and were still
On the Great Barrier adult P. im-
Reef of Australia, floating. Twenty hours after fertilization the embryos were
perialis are usually cryptic and nocturnal and are rarely hatched swimming gastrulae, still opaque yellow-tan in
encountered even on night dives. However, on the night color, and floating (Figs. 1, 5). It was not noted whether
of 20 November 1986, approximately 20 P. imperialis gastrulation occurred before hatching. Twenty-four hours
were collected by divers in an area less than 10 m in di- after fertilization there was considerable variation in the
ameter. This was 4 days after the full moon and the same morphology of the larvae, ranging from embryos that were
night as the mass spawning of acroporid scleractinian cor- nearly rectangular with one pair of arm rudiments (Fig.
URCHIN LARVAL DEVELOPMENT 79
Table I larva, a stomodeal invagination (larval mouth) never
schedule oJ’Phyllacanthus imperialis formed. In histological sections of gastrulae, large numbers
of intracellular yolk-like granular inclusions can be seen
Time since fertilization Stage within the mesenchymal, endodermal, and ectodermal
cells of the embryo (Figs. 5,6). The cells of the embryonic
30 min First cleavage
IO h Ciliated blastula wall and the archenteron are distinctly columnar, with
24 h Swimming gastrula, some 2 arms their nuclei arranged distally within the cells. The re-
48 h 4 short arms mainder of the cellular inclusions other than yolk-like
12 h Arms elongating, rudiment visible bodies are also located near the nucleus; the yolk-like
96 h Attachment bodies are stacked into columns that are generally directed
100 h Arms break off
116 h Fully formed juvenile toward the center of the gastrula (Figs. 5, 6). The blasto-
pore is unusually wide, with the opening (110 pm) being
almost one-quarter the diameter of the entire embryo
2) to larvae that were beginning to develop four equal- Histological cross sections through the preoral lobe of
length arms (as seen at 2 days old; Fig. 3). A few of these early plutei show the hydropore opening on the dorsal
larvae were swimming just below the surface of the water. surface of the larva, and that the hydrocoel, juvenile gut,
The first evidence of development of the echinus rudiment and perivisceral coelom have all formed (Fig. 7; compare
was apparent within the oral field of these early four-armed Fig. 15a). Cross sections taken near the posterior of these
plutei (Fig. 3). A band of cilia, extending along the margins larvae show the extent of development of the water vas-
of the arms and coursing up onto the preoral lobe, can cular system with the lumens of the primordial tube feet
be seen to be just beginning to form on these larvae (Fig. having formed. Except for the perivisceral coelom, all
3). Three days after fertilization four-armed plutei with coelomic spaces are filled with cellular and yolky-vesicle
well-developed echinus rudiments (Fig. 4) were distributed inclusions (Figs. 7, 8, 9, 10). The epidermis of the larva
throughout the culture vessel. Purple pigment granules is a single layer of cells nearly cuboidal in form.
appeared around the preoral lobe of the larvae and along
the arms. Many larvae 3 days old had begun to settle on Three- and four-day-old larvae
the bottom and sides of the culture beaker. At this point
virtually no larvae remained at the surface of the water. At 3 days the larva is fully formed (Fig. 4), and the
Six days after fertilization all larvae had settled and meta- juvenile rudiment is very conspicuous. A single contin-
morphosed into juvenile urchins. These juveniles were uous ciliary band runs up and down the length of each
maintained for another 3 weeks in the beaker without arm. At the juncture of the posterodorsal arms the ciliary
further care (changing of water or feeding). band courses up onto what appears to be the vestige of
the preoral lobe (Fig. 11). The preoral lobe has undergone
Description of embryology a torsion of nearly 45” to the left, as evidenced by the
twisting of the ciliary band on the preoral lobe (Fig. 1 I),
Details of the first few hours of development were not and the ciliary band disappears along the distal edge of
observed, but observations on fixed embryos revealed that the preoral lobe (Fig. 11). Oblique sections though 3-day
there was no wrinkling of the blastula or gastrula. Ap- plutei show an extensive convoluted juvenile gut and fully
proximately 15 h after fertilization a typical smooth gas- formed tube feet (Figs. 9, 10). Juvenile spines and pedi-
trula was formed by invagination of the blastular wall at cellaria are also present (Figs. 4, 13, and 14), giving the
the vegetal pole (Fig. 5). Histological sections of gastrulae, impression that 3-day-old larvae are in reality juvenile
20 h after fertilization, are reminiscent of the typical gas- rudiments with somewhat shortened larval arms. This is
trula of feeding echinoid larvae. The ectodermal wall is similar to the larvae of Eucidaris thoursi as described by
relatively thin and there is a distinct blastocoel surround- Emlet ( 1988).
ing the archenteron. At this time the archenteron extends The juvenile rudiment, with five primary podia, de-
approximately one-third the length of the blastocoel with velops out of the left side of the larva (Figs. 4, 11, 12, 13).
the distal tip bending toward the ventral surface of the There is no evidence of a juvenile mouth on the oral sur-
embryo at an angle of approximately 45” (Fig. 5). The face of the rudiment (Fig. 12). The posterior of the larva
timing of mesenchyme migration into the blastocoel was shows a striking bilateral symmetry (Fig. 14). With cross-
not observed, but in embryos 20 h old, a large clump of polarized light, the larval skeleton was observed to extend
mesenchyme cells was aggregated around the tip of arch- to the tips of the arms (Fig. 16). The extent of development
enteron (Figs. 1, 5) and around its base (Fig. 5). Though of the larval skeleton within the body of the larva was
the archenteron bends toward the ventral surface of the obscured by the development of juvenile skeletal struc-
80 R. R. OLSON ET AL.
Figure 1. Scanning electron micrograph of a fractured gastrula of Phyllacanthus imperialis showing
mesenchyme cells (ME) aggregated around the base of the archenteron. Note the small blastocoel (BC).
Figure 2. One-day-old embryo showing rudiments of post-oral arms (POA) and blastopore (B).
Figure 3. Two-day-old larva. PDA, posterodorsal arms; TF, tube foot; POL, preoral lobe.
Figure 4. Anterior view of 3- to 4-day-old larva. The preoral lobe has undergone torsion nearly 45” to
the left from its orientation in a typical pluteus. CB, ciliary band.
tures and the general opacity of the preoral lobe. Scanning vestibule, which is an invagination that forms on the left
electron micrographs of arm rods showed them to be fen- side of the larva. In cidaroids, there is no vestibule (Emlet,
estrated approximately 1 mm in length with a lattice-like 1988), but juvenile structures develop on the left side of
plate at the base (Fig. 17). The larvae of P. imperialis the echinopluteus in the same general location as on other
swim in a typical echinopluteus fashion with the anterior- echinoid larvae. The onset of metamorphosis can be ob-
posterior axis oriented vertically and the posterior directed served when juvenile structures are visible on the exterior
downwards much like the orientation of a falling bad- of an echinopluteus larva. These juvenile structures are
minton shuttlecock (Fig. 13). called the “echinus rudiment.”
Settlement and metamorphosis of P. imperialis larvae
Metamorphosis and description of juvenile urchin began 4 days after fertilization. Larvae settled by attaching
themselves to the side or bottom of the culture vessel with
Metamorphosis of the echinopluteus includes the for- their tube feet splaying their arms out radially. Within
mation of juvenile structures such as tube feet, spines, 4 h the tissue at the tips of the arms began to retract,
and pedicellaria, and the resorption of larval tissues. In exposing the ends of the skeletal rods. Six hours after at-
the majority of echinoids with an echinopluteus larvae, tachment the tissue on all arms pulled back to the main
the first appearance of juvenile structures is within the body of the larva. Over the next 4 h the four arm rods
URCHIN LARVAL DEVELOPMENT 81
broke off at their bases. Periodic upward jerking move- opment in echinoderms. First, egg size increases; second,
ment of the spines may have facilitated this breakage. typical pluteus structures are lost; and third, the appear-
Twenty hours after attachment all resemblance to the lar- ance of juvenile features is accelerated.
val form was lost and small juvenile urchins remained. The pattern of development observed in P. imperialis
Perhaps because the 4-day-old larva was virtually a fully is consistent with all three of these criteria. The eggs of
formed juvenile urchin, little morphological change ap- P. imperialis are larger than those of planktotrophs, and
pears to have occurred either internally or externally at some pluteal structures (such as the second and fourth
this point. Although numerous juveniles survived and pairs of arms and the mouth) do not develop. Finally
formed large numbers of spines, there was still no evidence there is the rapid (when compared to planktotrophic
of a mouth on urchins 2 1 days old. The nutritional needs forms) appearance of juvenile features. At 48 h after fer-
of the young urchins are probably met from remaining tilization, the juvenile gut is forming (Fig. 7), the primary
stored nutrients or possibly through uptake of dissolved podia are visible externally (Figs. 11, 12, 13) even though
organic matter; this is similar to the early development the larval arms have only just begun to develop, and ju-
reported for Heliocidaris erythrogramma (Williams and venile spines and pedicellaria are well formed.
Anderson, 1975). An important question in considering the apparent shift
from planktotrophy to direct development in echinoids
Discussion is whether the loss of pluteus features occurs before, at
the same time as, or after an increase in egg size and a
The larva of P. imperialis is unlike any previously de- loss of feeding ability. In P. imperialis many larval features
scribed echinoid larva. Its external form is surprisingly have been retained despite the increase in egg size and
similar to a feeding echinopluteus, yet it is completely loss of feeding ability. However, there is clear evidence
lecithotrophic. The larva most similar to P. imperialis is for the disappearance of other pluteus features. The ciliary
that of the Japanese sand dollar, Peronella japonica, which band, which is well developed along the arms, is greatly
develops from an egg of 276~ym diameter into an echin- reduced on the preoral lobe (Fig. 1 l), and not all pluteus
opluteus-like larva with two, three, or four arms (Okazaki arms form. With the loss of feeding, the generation of
and Dan, 1954). The larva of P. japonica lacks a preoral water currents around the oral region is presumably no
region and does not retain pluteal bilateral symmetry longer necessary; these currents are known to facilitate
(Okazaki and Dan, 1954; Mortensen, 192 1). Phyllacan- particle capture. It would be interesting to examine the
thusparvispinus, a congener also from Australia, has larger effectiveness and structure of the preoral cilia in Clypeaster
eggs (700 pm) that develop into little more than opaque rosaceus, which is a yolky facultative planktotroph that
spheres on which five primary podia appear just before arises from 280~pm eggs. Since it has no obligatory need
settlement (Parks et al., 1989). for particulate food material (although it does gain from
In a review of the evolution of direct development in feeding), are its preoral cilia reduced?
sea urchins, Raff ( 1987) identified four patterns of devel- It is unlikely that the pluteus form in P. imperialis is
opment that could describe the transition from the feeding secondarily derived from a brooded embryo (Strathmann,
pluteus to complete direct development (as in brooded 1974; however, see McEdward ( 1992) for an argument of
embryos). These patterns, which he loosely correlated with re-evolution of pelagic larval development in an asteroid).
egg size, are (1) typical feeding pluteus (100 pm egg), (2) The retention of pluteus-like features in the larvae of
partial pluteus, non-feeding (300 pm egg), (3) direct de- P. imperialis may be insignificant from the perspective of
velopment with a floating larva (500 ym egg), and (4) the evolution of a form of lecithotrophic direct develop-
brooded by mother, complete direct development ( 1300 ment. More likely, the pattern of development displayed
pm egg). by P. imperialis provides insight into the transition from
The larva of P. imperialis has features of both groups a feeding larva to a nonfeeding larva and into the relative
2 and 3. It still has the echinopluteus form, placing it in importance of certain larval characters in making such a
group 2, but the buoyancy of the larva and size of the egg transition.
place it closer to group 3. Overall, it seems to be a more In addition to its unique development pattern, P. im-
reduced larval form than Peronellajaponica (Okazaki and perialis also shows an interesting pattern of spawning be-
Dan, 1954) in that it is more yolky; however, it is not as havior. At least at Lizard Island, this species spawns and
reduced as Asthenosoma ijimai (Amemiya and Emlet, the larvae develop coincidentally with the scleractinian
1992) which shows only the slightest of echinopluteus corals of the Great Barrier Reef. In this region, more than
traits (i.e., little more than primary tube feet and two pairs 100 species of scleractinian corals mass spawn over the
of “para-arms”). course of three nights between the full and last-quarter
Raff (1988) suggested that three major developmental moons in late spring (Babcock et al., 1986). These are
changes occur in the shift to lecithotrophic direct devel- joined by many species of other taxa such as polychaetes
82 R. R. OLSON ET AL
Figure 5. Light micrograph (LM) section of the gastrula of P. imperialis. Note blastopore (BP) and
Figure 6. LM close-up of the tip of the archenteron (AE), showing migration of yolk-filled mesenchyme
cells into the blastocoel (BC).
URCHIN LARVAL DEVELOPMENT 83
Figure Il. Dorsal view of 3- to 4&y-old larva. Note ciliary band which diminishes as it extends up
onto the preoral Iok.
Figure 12. View of the “oral region” of the juvenile rudiment from a 3day-old larva showing absence
Fimre 13. Swimming orientation of 3- and 4-day-old larva,
Figwe 14. Posterior view of a larva that is ready to settle. Note the presence of juvenile spines and
(PA Hutchings, pers. comm,) and holothurians (RR0 pers. larvae are initially quite buoyant and develop near the
observation). In the two years that we observed P. im- surface of the water.
perialis, its spawning coincided with that of the corals, It would be easy to speculate that direct development
and similar to corals, it settled out of the water column as observed in P. imperia/is could be a response to selective
4 to 6 days after fertilization (Babcock and Hey-ward, pressures that have caused this species and corals to eon-
1986). Additionally, both P. imperiah larvae and coral verge on a similar pattern of development. However, direct
Figure 7, LM cross section through the preoral lobz of 2-&y-old larva. HY, hydropore; HC, hydroccel;
G, gut. See Figure I5a for reference.
Figure 8. LM cross section through the body of a Z-day-old larva. FOA, preoral arm: G, gut; TF, tube
foot, See Figure 1Sa for reference.
Figure 9. LM sagittal section through the ptebral lobe of a J-day-old larva. See Figure I5b for r&rem%.
Figure 10, LM rag&al section through the preoral lobe and arms of a 3day-ald larva. See Figure 15b
84 R. R. OLSON ET AL.
Figure 15. Diagrammatic representation of locations of light micrograph sections shown in Figure 7
(a:a’), Figure 8 (bb’), Figure 9 (c:c’) and Figure 10 (dd’).
lecithotrophic development in echinoids is not limited to throgramma (Williams and Anderson, 1975) produce lar-
the Great Barrier Reef, or even to tropical regions. Phyl- vae that are ecologically similar to P. imperialis, yet these
lacanthus parvispinus (Raff, 1987) and Heliocidaris ery- species live in temperate habitats where the corals are not
It is difficult to determine exactly why a species un-
dergoes mass spawning (Babcock et al., 1986); however,
further examination of the spawning behavior of P. im-
perialis across a broader geographic region might show
patterns that point more clearly to the exact environ-
mental cues that trigger such behavior. This type of study
might also show whether the pattern of development ex-
hibits any geographic variation that could provide insights
into the evolutionary reasons for an invertebrate to shift
its pattern of development from feeding to nonfeeding.
This project was supported in part by a Harbor Branch
postdoctoral fellowship. We thank the Lizard Island Re-
search Station for providing field facilities and the Aus-
tralian Institute of Marine Science for logistical support.
Field assistance was provided by A. R. Davis, E. M. Ley-
decker, R. Z. McPherson, M. B. Olson, K. Osborne, L.
Sullivan, and P. Watts. Special thanks to P. Dixon for
videotaping metamorphosis and to P. Linley for assisting
with SEM. This paper has benefited from discussions with
R. Emlet and R. Raff. Harbor Branch Oceanographic In-
stitute Contribution #965. University of New Hampshire,
Center for Marine Biology Contribution #28 1.
Amemiya, S., and R. B. Emlet. 1992. The development and larval
form of an echinothurioid echinoid, Asthenosoma ijimai, revisited.
Biol. Bull. 182: 15-30.
Babcock, R. C., G. D. Bull, P. L. Harrison, A. J. Heyward, J. K. Oliver,
C. C. Wallace, land B. L. Willis. 1986. Synchronous spawnings of
Figure 16. Light micrograph of the larval skeleton of P. imperialis 105 scleractinian coral species on the Great Barrier Reef. Mar. Biol.
highlighted with cross-polarized light. 90: 379-394.
Figure 17. S.E.M. of a skeletal rod from one arm of a 4-day-old Babcock, R. C., and A. J. Heyward. 1986. Larval development of certain
larva of P. imperialis. gamete-spawning scleractinian corals. Coral Reefs 5: 11 l- 116.
URCHIN LARVAL DEVELOPMENT 85
Christiansen, F. B., and T. M. Fenchel. 1979. Evolution of marine Okazaki, K., and K. Dan. 1954. The metamorphosis of partial larvae
invertebrate reproductive patterns. Theor. Popul. Biol. 16: 267-282. of Peronellu juponicu Mortensen, a sand dollar. Biol. Bull. 106: 83-
Emlet, R. B. 1986. Facuhative planktotrophy in the tropical echinoid 99.
Clypeuster rosaceus (Linneaus) and a comparison with obligate Parks, A. L., B. A. Parr, J. Chin, D. S. Leaf, and R. A. Raff.
planktotrophy in Clypeuster suhdepressus (Gray) (Clypeasteroida: 1988. Molecular analysis of heterochronic changes in the evolution
Echinoidea). J. Exp. Mar. Biol. Ecol. 95: 183-202. of direct developing sea urchins. J. Evol. Biol. 1: 27-44.
Emlet, R. B. 1988. Larval form and metamorphosis of a “primitive” Parks, A. L., B. W. Bisgrove, G. A. Wray, and R. A. Raff. 1989. Direct
sea urchin, Euciduris thonursi (Echinodermata:Echinoidea:Cida- development in the sea urchin Phyllucunthuspurvispinus (Cidaroidea):
roida), with implications for developmental and phylogenetic studies. Phylogenetic history and functional modification. Biol. Bull. 177:
Biol. Bull. 174: 4-19. 96-109.
Paul, C. R. C., and A. B. Smith. 1984. The early radiation and phylogeny
Emlet, R. B. 1990. World patterns of developmental mode in echinoid
of echinoderms. Biol. Rev. 59: 443-48 I.
echinoderms. Pp. 329-335 in Advances in Invertebrute Reproduction, Raff, R. A. 1987. Constraint, flexibility, and phylogenetic history in
Vol. 5, M. Itoshi and 0. Yamashita, eds. Elsevier Science, Amsterdam. the evolution of direct development in sea urchins. Dev. Biol. 119:
Emlet, R. B., L. R. McEdward, and R. R. Strathmann. 1987. 6-19.
Echinoderm larval ecology viewed from the egg. Pp. 5% 136 in Echi- Raff, R. A. 1988. Direct developing sea urchins: a system for the study
noderm Studies, Vol. 2, M. Jangoux and J. M. Lawrence, eds. Bal- of developmental processes in evolution. Pp. 63-69 in Echinoderm
kema, Rotterdam. Biology, R. Burke, P. Mlandenov, P. Lambert, R. Parsley, eds. Bal-
McEdward, L. R. 1992. Morphology and development of a unique kema Press.
type of pelagic larva in the starfish Pteruster te.sse/utus(Echinoder- Strathmann, R. R. 1974. Introduction to function and adaptation in
mata: Asteroidea). Biol. Bull. 182: I77- 187. echinoderm larvae. Thulussia Jugos. 10: 321-339.
McMillan, W. O., R. A. Raff, and S. R. Palumhi. 1992. Population Strathmann, R. R. 1979. Echinoid larvae of the Pacific northeast (with
genetic consequences of developmental evolution in sea urchins (genus a key and comment on an unusual type of planktotrophic develop-
Heliociduris). Evolution 46: 1299- 13 12. ment). Can. J. Zool. 57: 610-616.
Vance, R. R. 1973. On reproductive strategies in marine benthic in-
Mortensen, T. 1921. Studies of the Development and Larval Forms of
vertebrates. Am. Nut. 107: 339-352.
Echinoderms. G.E.C. Gad., Copenhagen. Williams, D. H. C., and D. T. Anderson. 1975. The reproductive system,
Mortensen, T. 1938. Contributions to the study of the development embryonic development, larval development, and metamorphosis of
and larval forms of echinoderms IV. Kgl. Dan. Vidensk. Selsk., Skr. the sea urchin Heliociduris erythrogramma (Val.) (Echinoidea: Echi-
Naturvid. Math. Afd. (9), 7(3): l-59. nometridae). Aust. J. Zool. 23: 37 l-403.
Ohrehski, S. 1979. Larval colonizing strategies in marine benthic in- Wray, G. A., and R. A. Raff. 1991. The evolution of developmental
vertebrates. Mur. Ecol. Prog. Ser. 1: 293-300. strategy in marine invertebrates. Trends Rex Ecol. Evol. 6: 45-50.