BULLETIN OF MARINE SCIENCE, 76(1): 1–26, 2005

NEW FISSION PROCESSES IN THE ZOANTHID PALYTHOA

CARIBAEORUM: DESCRIPTION AND QUANTITATIVE ASPECTS

Alberto Acosta, Paul W. Sammarco, and Luiz F. Duarte

ABSTRACT
Populations of the zoanthid Palythoa caribaeorum (Duchassaing and Michelotti,
1861) were monitored in a 2-yr study at several sites on the São Paulo coast, Brazil,
to describe, quantify, and compare fission processes occurring in moderately vs.
highly stressed environments. The effects of depth and season were also consid-
ered. Palythoa caribaeorum was found to reproduce by two new general processes
of fission. The first process requires: A) the formation of crevices throughout the
colony and the maintenance of basal coenenchyme throughout; B) the production
of polyp-clusters which are connected to each other by thin basal coenenchyme, not
readily visible; and C) the subsequent severance of the basal coenenchyme between
polyp-clusters, creating true ramets. The first process included two variants: edge
fission and polyp-cluster release. The second fission process was characterized by
directional growth of tissue and a subsequent breaking away and dispersing of the
ramets. It also included two variants: tissue outgrowth and polyp-ball production.
Of the monitored colonies, 55% (nt = 579) exhibited at least one variant of fission,
yielding 1304 ramets over 1 yr. Edge fission was the dominant form of asexual repro-
duction in this species. We found no evidence to support the hypothesis that higher
environmental stress results in increased ramet production among sites, depths, or
seasons. Fission occurred year-round and in colonies of all sizes. The minimum size
of a colony reproducing by fission was 5 cm2. A significant positive linear relation-
ship was found between colony area and number of ramets produced per colony.
Ramet release also occurred year-round, although the frequency of release was sig-
nificantly lower during the winter, when environmental conditions were harsher.
This is the first report of tissue outgrowth, polyp-ball production, and polyp-cluster
release for the genus Palythoa, and the first report of polyp-ball production within
the Zoanthidea. There appear to be no analogues of polyp-cluster release within the
Cnidaria. Fission in P. caribaeorum appears to be a conservative trait over a wide
geographical range. It also seems to be endogenously controlled (genetically pro-
grammed) responding to colony growth constraints. Exogenous factors, however,
may help to define the variant of fission used and the quantity of ramets produced at
a given time. The adaptive value of fission for P. caribaeorum lies in its contribution
to the number of clone-mates (increasing population size). It represents a critical
and important form of asexual reproduction, helping to explain the ecological and
evolutionary success of this species in the western Atlantic.

Asexual reproduction is the process by which an increase in colony numbers is

achieved without the aid of genetic recombination. Anthozoa possess many repro-
ductive modes, which is consistent with the high degree of developmental plasticity
known to occur in the Cnidaria (Chia, 1976; Harrison and Wallace, 1990; Shostak,
1993; Ryland, 1997). New colonies can be formed asexually in several ways, includ-
ing polyp bail-out (Sammarco, 1982), coral polyp expulsion (Kramarsky et al., 1997),
detachment of a polyp-cluster or polyp-balls (Rosen and Taylor, 1969), fragmenta-
tion of colonies (see review in Highsmith, 1982), asexually produced planulae (Bell,
1982; Ayre and Resing, 1986), pedal laceration (Chia, 1976; Cooke, 1976; Shick, 1991),

Bulletin of Marine Science 1

© 2005 Rosenstiel School of Marine and Atmospheric Science
of the University of Miami
2 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

budding (Vaughan and Wells, 1943; Chia, 1976), and colony fission (McFadden, 1991;
Shick, 1991).
Fission is a primary mode of asexual reproduction in numerous anthozoans, in-
cluding anemones, zoanthids, soft corals, and scleractinian corals (e.g., Chia, 1976;
Hughes and Jackson, 1980; Karlson, 1991; Tanner, 1999). Fission may be manifested
via longitudinal fission (see Chia, 1976; Shick, 1991; Shostak, 1993), transverse fis-
sion (Crowell and Oates, 1980; Cairns, 1988; Soong et al., 1995), binary fission (Mc-
Fadden, 1986; Hughes, 1989), or autotomy (Walker and Bull, 1983; Hughes, 1989;
Dahan and Benayahu, 1997). It is necessary, however, to distinguish between colony
fission (McFadden, 1986), which contributes to population size, and fission of a uni-
tary polyp, which contributes to colony growth (polyp fission).
Reproductive strategies in the Zoanthidea have been reviewed recently by Ryland
(1997). Zoanthids are known to form clones and propagate by colony fission (Cooke,
1976; Yamazato and Isa, 1981; Karlson, 1991; Burnett et al., 1994; 1995; Soong et al.,
1995; Tanner, 1999), transverse division (Soong et al., 1995), colony fragmentation
(Karlson, 1986; 1988a,b; 1991; Ryland, 1997; Acosta et al., 2001), and budding (Larson
and Larson, 1982; Ryland, 1997). No detailed description currently exists regarding
the fission process at the colony level in Palythoa (also see Yamazato and Isa, 1981;
Ryland, 1997; Tanner, 1997). In addition, it is not known whether species of Palythoa
from the western Atlantic reproduce by fission at all, and, if so, how.
Here we describe the morphological characters associated with colony fission for
Palythoa caribaeorum (Duchassaing and Michelotti, 1861) and their development
throughout the process. We also describe the different means by which ramets may
be produced in the field. We define fission as endogenous division of the colony into
portions that regenerate into separate individuals (Hughes, 1989; Karlson, 1991;
Acosta et al., 2001). Control of fission in the Zoanthidea is hypothesized to be en-
dogenous (genetically programmed), regulating stolonal elongation, budding of new
polyps, and stolonal degeneration (Sebens, 1979; Muirhead and Ryland, 1985; Karl-
son, 1991). Environmental stress may also control fission in anthozoans (Shick and
Hoffmann, 1980; Karlson, 1983; Tanner, 1999). The matter of whether fission in the
Cnidaria is triggered and regulated by endogenous or exogenous factors, or both re-
mains controversial. Some investigators have suggested that an increase in fission
rate can represent a response to physical stress and partial colony mortality (Harper,
1977; Cook, 1978; Shick and Hoffmann, 1980; Shostak, 1993). In this study, we also
attempt to determine whether there are significant differences in fission rate be-
tween populations of P. caribaeorum in two habitats characterized by different levels
of stress (i.e., under different light and sedimentation regimes; see Acosta, 2001 for
details), also with respect to different depths and seasons. The relationship between
ramet production and partial mortality was also considered.
Data related to seasonal variation in formation, size, and release of ramets, and to
the role of fission in the population size has been reported for many species, includ-
ing zoanthids (see Karlson, 1986, 1991; Dahan and Benayahu, 1997; Ryland, 1997;
Tanner, 1999). To date, however, no data have been published on any of these popula-
tion characteristics for P. caribaeorum, particularly regarding how new ramets pro-
duced via fission affect population growth.
Palythoa caribaeorum is a sessile, epibenthic, colonial organism, common on shal-
low reefs from the intertidal to 12 m depth (Sebens, 1982a). Colonies have been ob-
served to cover tens to hundreds of square meters in reef crest and reef flat zones in
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 3

the Caribbean and Brazil (Fadlallah et al., 1984; Gleibs, 1994; Migotto, 1997; Díaz
et al., 2000). It is a hermaphroditic broadcaster (Fadlallah et al., 1984; Gleibs, 1994).
High abundances of the zoanthid P. caribaeorum in the western Atlantic may be
explained by several factors (see Cooke, 1976; Suchanek and Green, 1981; Sebens,
1982a), including its reproductive capabilities (both sexual and asexual; Fadlallah
et al., 1984; Acosta and Asbahr, 2000; Acosta et al., 2001). Here, we quantify the
various processes by which fission is achieved, and examine how ramet production
may contribute to the abundance of P. caribaeorum on shallow reefs. Fission in this
species was considered at the colony level and not the cellular, tissue, or physiologi-
cal levels.

METHODS

SAMPLING PERIODS.—This study was performed over a 2-yr period. The first observations
were made from May 1996–May 1997. Three hundred colonies of P. caribaeorum were fol-
lowed along the São Sebastião channel, São Paulo, Brazil. Colonies were selected at two sites
and at two depths. One plot (6–9 m2 in size) with approximately 75 colonies was chosen
arbitrarily at each site and depth (n = 4 plots). The two study sites were Praia Portinho (PPo),
located on São Sebastião Island, and Ponta Recife (PR), on the continental coast (Fig. 1). There,
sites are near the southern latitudinal limit of geographical distribution of P. caribaeorum
(23ºS). The two sub-tidal depths sampled were 0.5–1.5 m, and 2.5–4.0 m, representing differ-
ent environmental conditions for the organism. Observations were made bimonthly to exam-
ine general patterns of fission, identify and describe different fission processes, and identify
characters associated with the process which could be quantified through space (different
habitats) and time (seasons). These data indicated that quarterly sampling would be sufficient
to accurately document, quantify, and compare all of the variants of fission observed.
EXPANDED SAMPLING.—Additional colonies of P. caribaeorum were monitored during the
second half of the study, from May 1997–May 1998. Three replicate plots were established at
each depth and site (6 plots site-1; 12 total plots), with 38–90 colonies plot-1. Plots ranged from
3.2 to 11.6 m2 in size, and were demarcated via steel nails driven into the bottom. In total, 579
colonies of various sizes (size range: 2.5 × 10 -5 to 2.1 m2) were monitored by sampling every
3 mo near the end of each season —August 1997 (winter), November 1997 (spring), February
1998 (summer), and May 1998 (autumn). Ponta Recife and the deeper habitats represented the
more stressful set of conditions (e.g., low light intensity and occasional higher sedimentation
rates), when compared to Praia Portinho and the shallower habitats, respectively (see Acosta,
2001 for details regarding statistical comparison of abiotic variables among sites, depths, and
seasons). Abiotic factors, such as storm frequency, also varied among seasons, particularly
in summer vs. winter (ibid). Data on oceanographic variables along with benthic community
structure associated with the sites, as well as additional information on P. caribaeorum colo-
nies sampled within the plots (e.g., percent-cover, colony density, etc.) may be found in Acosta
(2001) and Acosta et al. (2001).
Palythoa caribaeorum colonies were mapped underwater using acrylic slates and colony
size was measured in situ (max colony length × max width). This information facilitated the
relocation of specific colonies in the field, (these data were not used to estimate colony growth
rates). Individual colonies and polyp-clusters were followed through time. The number, area,
and position of ramets and their precursor polyp-clusters were measured (area = length ×
width), recorded, and monitored to help follow ongoing fission processes.
Outgrowths of tissue hanging at a colony’s edge (i.e., vertical side of a boulder) were also
followed closely for 3 mo (per colony). Position, size, and number of individual outgrowths
associated with a given colony were recorded, as were polyp-clusters growing upward from
the upper surface.
4 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Figure 1. Study sites in the São Sebastião channel, São Paulo coast, Brazil. (1) Ponta Recife, on
the mainland coast; and (2) Praia Portinho, on an offshore island.

A number of morphological characters were monitored to help describe the fission process
and its variants through time. The dynamics of the fission processes, ramet production, and
ramet release were assessed using the following variables: 1) Colony size (max colony length ×
max width; in cm2); a colony was defined as any group of polyps physically independent from
other groups, irrespective of size or proximity to a neighboring group; 2) Cumulative length
of actively forming crevices (gaps); crevices occurring externally in the colony surface and
spreading through the colony were measured with a cloth measuring tape and were moni-
tored through time; fully developed crevices defining established polyp clusters were not in-
cluded; 3) Number and area of polyp-clusters (precursors of ramets); “Polyp-cluster” refers
to a group of polyps within a colony that are visually distinct, but connected to each other
and the surrounding parent colony by a thin basal coenenchyme (see Karlson, 1986; Ryland,
1997). Basal coenenchyme was defined as an extension of the column (body walls), with com-
mon tissue connecting the polyps or filling the spaces between polyps in P. caribaeorum (see
Verrill, 1869; Duerden, 1898; Calgren, 1937); 4) Number and area of outgrowths of tissue;
5) Time required for ramet formation; defined as the period of development from initiation
of the fission process or of the study (t0) to complete development of a ramet. A ramet was
defined as a group of polyps physically disconnected from the parent colony “a physiologi-
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 5

cally independent unit, or separated vegetative part of the genet” (Harper, 1977) through the
absence of connecting basal coenenchyme. This was verified visually by prying the polyp-
clusters apart. Ramet release was defined as detachment of the ramet from the substratum,
and is not to be confused with simple separation from the parent clone (ramet production); 6)
Number of ramets produced per colony, and the process of fission by which they were produced
(see Fig. 2); 7) Ramet area; This allowed us to determine the mean and variance of the size
of ramets produced by the different variants of fission, respectively. 8) Temporal variation of
ramet production; measured to determine seasonal variation in the processes of fission and
ramet detachment (see Acosta et al., 2001 for definitions); 9) Area of partial mortality; partial
mortality was considered to be any portion of the colony exhibiting lesions or skeletal rem-
nants (e.g., caused by disease; Acosta, 2001). The area of partial mortality was calculated as
the diameter of the affected area (or, in some cases, max length × width).
Population growth is defined here as the potential change in population size with time
(Akcakaya et al., 1997), in our case, as a result of new ramets created by fission that may or
may not increase the total number of colonies in a given population. It does not refer to colony
growth (increase in polyp number) or population cover (total number of zooids, or existing
mass).
DATA ANALYSIS.—As all variables were non-normally distributed, data were transformed
(using log10 x or log10 x + 0.5) prior to analysis by parametric statistics (Sokal and Rohlf, 1995).
Bartlett’s and Fmax Tests for Homogeneity of Variance were performed to insure homosce-
dasticity. Data were analyzed in space and time by one-way ANOVA, followed by a posteriori
tests using the Tukey-Kramer Test (Sokal and Rohlf, 1995). Two-way ANOVAs were used
to compare several variables related to fission (site × depth). The total number of colonies
exhibiting a particular type of fission was compared by G-test using Williams’ correction
(Sokal and Rohlf, 1995). This was also done for the number of ramets produced by each type of
fission, and the frequency of colonies requiring 3, 6, 9, or 12 mo to produce ramets. Seasonal
changes in ramet detachment were also examined. The relationship between colony area and
number of ramets produced per colony was analyzed using a Model-II least-squares linear
regression analysis (Sokal and Rohlf, 1995).

RESULTS

FISSION PROCESSES.—Two general processes of fission were identified in P. carib-

aeorum, and each had two variants (Table 1). We defined these four variants as edge
fission, polyp-cluster release, tissue outgrowth, and polyp ball production (Fig. 2).
The first general fission process was always characterized by a previously existing
crevice and the formation of a polyp-group (see below); the second general process
was not. This represents a first description for these fission processes in this species.
The terms are derived from relevant morphological characters associated with each
in the field. These processes and their variants were not part of a unidirectional,
sequential process (Table 1). Rather, they were alternative (independent and non-se-
quential) means by which to achieve a common end —ramet production.
In the first fission process, which requires crevice formation as a preliminary step,
clearly visible crevices appear in one or several parts of the colony. They progress
linearly at first, and then branch throughout the colony. They spread between pol-
yps, dividing the parent colony into one to tens or even hundreds of polyp-clusters,
depending on colony size (Fig. 2A). Each polyp-cluster is still physically inter-con-
nected aborally to neighboring groups and the parent colony by a thin, flat, basal
coenenchyme.
Basal coenenchyme was the result of vertical constriction of coenenchyme connect-
ing the polyps and was not visible without parting the overlying polyps. Subsequent
6 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Figure 2. Fission processes of Palythoa caribaeorum. Fission Process I: (A) Colony exhibiting
crevices spreading throughout the colony surface (arrow), dividing a colony into several polyp-
clusters. Colony thickness = 4.5 cm. (B) Edge Fission. Parental colony (upper center) still con-
nected by basal coenenchyme to a polyp-cluster (arrow left) at the colony edge. (C) Parent colony
and a new physically isolated, smaller ramet (arrow); P. variabilis polyp (at bottom right). Polyps
in colonies b and c are 1.0–1.5 cm in height. (D) Polyp-cluster Release. One polyp-cluster (arrow)
is becoming an independent ramet via the loss of connective basal coenenchyme to the neighbor-
ing groups and simultaneous detachment from the substratum; colony thickness = 4.0 cm. Fission
Process II: (E) Outgrowth. Tissue growing beyond the substratum; two different colonies are
shown (arrows), each exhibiting tissue outgrowth at their edges (note teardrop shape), detached
from the substratum. (F) Polyp-ball production. A group of ≤12 polyps growing upward from
the colony surface (arrow). The polyp ball remaining connected to neighboring polyps by distal
coenenchyme, but detached from the substratum. Polyp diameter = 0.8 cm
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 7

disappearance of the coenenchyme (probably through resorption or degeneration)

progressed from the oral to the aboral portion of the polyps, leaving only a small por-
tion of tissue connecting the polyps. The remaining tissue was 1–3 mm high, leaving
a gap of 1–10 mm between polyp-clusters. Visually detectable crevices are the result
of coenenchyme digestion.
Crevices formed throughout the year and were a key character indicating active
fission. Crevices generally continued to form in large colonies until most of the polyp-
clusters were about the same size. The parent colony may have been regulating polyp-
cluster (ramet precursor) size. Eventually the fission process was completed (ramet
formation). Ramets were formed when the basal connecting tissue surrounding a
given polyp-cluster became progressively thinner, eventually breaking, making the
new ramet physically and physiologically independent of the parent colony (Fig. 2C).
All polyp-clusters are capable of becoming autonomous ramets, although this pro-
cess can require several weeks to > 2 yrs. This is in part because the parent colony
can freeze the fission process at any stage. New ramets formed by fission sometimes
remained attached to the substratum, alongside the parental colony (Fig. 2C), later
becoming detached from the substratum to disperse elsewhere. The variants of edge
fission and polyp-cluster release fall within the first type of fission described above
(Type I). Here, ramets must pass through the stage of being polyp-clusters.
VARIANT I-A, EDGE FISSION.—Polyp-clusters occurring at the edge of the colony
were usually thin, with small polyps (Fig. 2C), and while actively growing, were often
involved in competition for space (Table 1). The ramets formed from polyp clusters at
the colony edge bore 3 to ≥ 20 polyps and were ~0.5 mm–1.5 cm in height. They gen-
erally remained attached to the substratum, within mm to cm of the parent colony,
for 3–6 mo. Parent colonies were capable of producing dozens of ramets simultane-
ously in this way (Table 2).
In many cases, colonies utilizing edge fission grew to monopolize all nearby avail-
able substratum (i.e., boulders). Ramet formation was infrequent where there was ac-
tive, intra-specific (inter-genet or -ramet) competition for space, but frequent under
conditions of inter-specific competition for space, particularly with such epibiota as
Palythoa variabilis Duerden, 1898 (Fig. 2C), Zoanthus sociatus (Ellis and Solander,
1786), Zoanthus solanderi Leseur, 1817, several species of ascidians, sponges, Sargas-
sum sp., and other macroalgae.
VARIANT I-B, POLYP-CLUSTER RELEASE.—The release of polyp-clusters, also pre-
ceded by crevice formation and fission (Fig. 2A,D), involved lateral separation of the
cluster from neighboring clusters as it simultaneously separated itself aborally from
the substratum (Fig. 2D). The ramet would then detach from the substratum, drift
away, and reattach elsewhere. Colonies releasing polyp-clusters were always large in
size, bearing a high number of polyp-clusters (max 400 per colony) and possessing
long polyps (2.5–6.0 cm in height; Fig. 2D). A ramet produced by this process always
originated near the center of the colony.
Parent colonies utilizing this variant of fission were substrate-limited, having mo-
nopolized all local substratum available (Table 1). These colonies could produce sev-
eral ramets by this or any other variant of fission simultaneously.
Ramets formed by edge fission or polyp-cluster release were never observed to fuse
again with the parent colony. This irreversible developmental phenomenon pertains
to both crevice formation and the formation of polyp-clusters.
In the second process of fission (Type II), we found two variants termed tissue
 8

Table 1. Comparison of traits characterizing and differentiating the two major processes and four variants of fission found in Palythoa caribaeorum.

Process/variants Edge fission Polyp-cluster release Tissue overgrowth Polyp-ball production

Fission process: I
Crevice formation followed by fission Yes (obligate) Yes (obligate) No No
Ramet precursor Polyp-cluster Polyp-cluster
Position in colony Edge Center
Attachment of ramet precursor to substratum Yes Yes
Separation of ramet from substratum No Yes
Fission (breakage) Basal coenenchyme Basal coenenchyme
Fission process: II
Directional growth followed by fission No No Yes Yes
Shape Teardrop-shaped Ball-shaped
extension of tissue
Position within colony Edge Near center
Attachment to substratum Hanging from edge Attached only to
of substratum neighboring polyps
Mechanism of fission Autotomy, or broken by Distal coenenchyme
extrinsic factors
Frequency of fission Frequent Common Common Rare
Detachment of ramet May remain beside parent for Detachment is almost Detachment is immediate Detachment is immediate
weeks or months before detach- immediate
ment
BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Relative size of ramet Medium Large Small Small

Polyp size within ramet Small Large (> 2.5–6 cm) Small Normal (1–1.5 cm)
Exogenous factors associated with fission
Substrate limitation Yes, at times Yes Yes Yes, at times
Light limitation Yes No Yes No
Interspecific competition for space Yes No Yes No
Intraspecific competition for space No No No No
Substrate irregularity (e.g., shells) No No No Yes
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 9

outgrowth and polyp-ball production (Fig. 2E,F). Here, crevice formation was not a
prerequisite for ramet formation; instead, tissue growth occurred at specific points
within the colony over a period of weeks–months before those polyps became physi-
cally independent ramets.
VARIANT II-A, TISSUE OUTGROWTH.—Tissue outgrowth occurred exclusively at
the colony edge (Fig. 2E) on the tops of large boulders (20–150 cm from the bottom),
where no suitable substratum was available for lateral growth. Under these condi-
tions, polyp growth was directed (via budding) towards specific points along the
colony edge. One to several portions of unattached tissue per colony continued to
grow, hanging from the colony edge (Fig. 2E). These outgrowths were generally small
in size and teardrop-shaped, reaching lengths of 1–100 cm and widths of 2–8 cm. Be-
cause of their position, they were usually exposed to low light levels and sometimes
inter-specific competition for space with other sessile epibenthic organisms (Table
1). Eventually, the formation broke away from the parent colony by autotomy (i.e.,
tissue degeneration of the thinnest portion of the tissue), its own weight, or external
forces (e.g., currents or storms). Ramet formation was then completed.
Tissue outgrowth did not always result in ramet formation. Sometimes the ex-
tended tissue continued to grow (> 1 m), reattaching to the bottom and expanding
the parent colony.
VARIANT II-B, POLYP BALL PRODUCTION.—Polyp ball production varied from tis-
sue outgrowth in that a large single polyp initially grew upward from the colony
surface and then budded into a “ball shape” (Fig. 2F). A resulting small group of
polyps (≤ 12) then emerged from the colony surface (similar to that described for
the Scleractinia; Rosen and Taylor, 1969; Dahan and Benayahu, 1997). The group
was not attached to the substratum but to neighboring polyps via coenenchyme (Fig.
2F). Eventually, the connecting coenenchyme degenerated, freeing the polyp-ball to
be carried away by currents. Polyp balls were formed generally at the interior of the
colony.
FREQUENCY OF FISSION.—Fission occurred in 55% of the 579 colonies monitored
over 1 yr. An additional 19.2% of the population exhibited crevice formation as well
as an increase in the number of polyp-clusters, bringing the total number of colonies
actively involved in fission to 74.2% (n = 320). Of these, 47.8%–74.1% reproduced
asexually by more than one variant of fission (Table 2).
There were a total of 1304 ramets produced and released within 1 yr (Fig. 4), which
was almost twice the number of polyp clusters (767) produced over the same period.
This was in addition to the 1772 polyp-clusters already present in the population at
the beginning of the study, representing an extraordinarily high fission rate.
QUANTITATIVE COMPARISONS.—The frequency of different variants of fission and
the mean number of ramets produced per colony by these variants were significantly
different (P < 0.001, G-test with Williams’ correction; P < 0.001, one-way ANOVA).
Edge fission produced the highest number of ramets per colony (72.7%; P < 0.05,
Tukey-Kramer Test), followed by polyp-cluster release (20.5%) and tissue outgrowth
(6.8%; Fig. 4).
Colonies utilizing these three types of fission varied significantly from each other
with respect to a number of different characters (mean values; Fig. 5). These includ-
ed: 1) cumulative crevice length colony-1 (P < 0.001, one-way ANOVA); 2) number
of polyp-clusters colony-1 (P < 0.05); 3) area of partial mortality colony-1 (P < 0.001);
and 4) size of parent colony (P < 0.05). Fission by tissue outgrowth had significantly
 10

Table 2. Comparison of abundance and frequencies of Palythoa caribaeorum colonies exhibiting various characters associated with edge fission, polyp-cluster
release, and tissue outgrowth. Note that any given colony many reproduce by one or more variants of fission simultaneously; thus, totals may be disparate. “Es-
timated Cumulative Area of Ramets” was calculated based upon 1,304 ramets known to be released over the period of a year, and also from the size-frequency
distribution data calculated from a subset of 524 ramets measured in the field.

Colony characteristics Edge fission Polyp-cluster Tissue Total

release overgrowth
Number of colonies exhibiting a given variant of fission (n) 247 137 58
Number of colonies exhibiting only this variant 129 48 15
Percent of colonies exhibiting only this variant 52.2% 35.0% 25.9%
Number of colonies reproducing by this plus at least one other variant 118 89 43
Percent of colonies reproducing by this plus at least one other variant 47.8% 65.0% 74.1%
Fission process – I, descriptive characters when active
Number of colonies exhibiting an increase in crevice length throughout the year 195 102 54
Number of colonies exhibiting an increase in polyp-cluster throughout the year 209 132 49
Maximum number of polyp-clusters produced by a single colony within 1 yr 162
Ramet characteristics and production
Maximum number of ramets produced by a single colony within 1 yr 77 18 6
Mean number of ramets produced per colony per year; (s.d; n) 3.8 (6.99; 247) 1.9 (2.04; 137) 1.4 (0.88; 58)
Mean ramet size, in cm2 (s.d., n) 32.2 (69.31; 276) 43.3 (140.00; 186) 22.4 (36.79; 62)
Range of ramet sizes 0.25–725 cm2 9.0–1,518 cm2 1.5–192 cm2
Area of colonies, mortality, and ramets produced
BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Mean colony size (s.d.) (n = 579) 1,414 (2,563) cm2

Sum of the area for all colonies sampled (n = 579) 82.5 m2
Sum of the area for all colonies exhibiting this variant (n = 320) 59.45 m2 33.22 m2 28.74 m2 72.2 m2*
Sum of the total area of partial colony mortality over 1 yr 2.37 m2 1.1 m2 1.53 m2 2.7 m2*
Percent of area with partial colony mortality for all colonies 4.0% 3.3% 5.3% 3.7%
Estimated cumulative area of ramets produced by each mode 3.05 m2 1.13 m2 1.88 m2 6.06 m2
Percent of total colony area released as ramets 5.1% 3.4% 6.5% 8.3%
* Note: A colony may exhibit one or more variants of fission simultaneously; thus, totals cannot be easily reconciled for all variants.
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 11

Figure 3. Example of the dynamics of fission in a typical colony of Palythoa caribaeorum, sam-
pled every 3 mo for 18 mo. The number of new ramets formed by edge fission, tissue outgrowth,
and polyp-cluster release is shown graphically. Number of polyp clusters present and ramets
released during any one period is also shown. The colony starts at t0 with 22 polyp-clusters, and
two ramets produced by edge fission. After 18 mo, this colony has 21 polyp-clusters (six of which
are new), one tissue outgrowth formation, and has produced and released ten ramets –five of them
by edge fission, four by polyp-cluster release, and one by tissue outgrowth. When crevices meet,
a polyp-cluster is formed; ramets are formed from a polyp-cluster separating from the parent
colony. Note that the number of polyp-clusters decreases with ramet formation and increases
when crevices meet.

higher levels of all of the above characters than the other two variants (P < 0.05,
Tukey-Kramer Test; see Table 2 for comparisons).
TEMPORAL VARIATION.—There was a constant rate of ramet production. In ad-
dition, colonies under different habitat conditions did not vary significantly in the
amount of time they required to develop a ramet (3 to > 12 mo; P > 0.05, G-test with
Williams’ correction). Once a ramet was fully developed, however, a significantly
greater proportion of colonies released the ramets (dispersing them) within 3 mo vs.
longer periods of time (P < 0.001, G-test with Williams’ correction; Fig. 6A). Ramet
release varied significantly with time, being lowest in the winter (Fig. 6B; P < 0.001,
G-test with Williams’ correction). No seasonal variation was detected in mean cu-
mulative crevice length, mean number of polyp-clusters, or mean ramet production
per colony (P > 0.05, one-way ANOVA, n = 320). In most cases, currents and wave
action were the principal forces of dispersal for ramets.
VARIATION BETWEEN SITES AND DEPTHS.—The number of polyp-clusters per col-
ony varied significantly with respect to depth (P = 0.01, one-way ANOVA; Table 3),
with higher numbers in deeper vs. shallow water. The number of polyp-clusters per
colony also varied significantly with respect to habitat (each site/depth vs. the others;
P < 0.05, one-way ANOVA), with a higher number in deeper water (7.9 ± 14.8, n =
153) than in shallow water (5.3 ± 7.2, n = 167; P < 0.05, Tukey-Kramer Test). Num-
ber of polyp-clusters per colony was highest at the PR-deeper site, and successively
lower in the PPo-deeper, PPo-shallow, and PR-shallow sites (P < 0.05, Tukey-Kramer
12 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Figure 4. Total number of ramets produced by a given form of fission in Palythoa caribaeorum
– edge-fission, polyp-cluster release, and tissue outgrowth, respectively. Any given colony may
reproduce by one or more variants of fission simultaneously. ni = 320 (total number of parent
colonies exhibiting fission); 1304 (total number of ramets produced). Polyp-ball production is a
rare event and is not shown here.

Test). No significant variation was found, however, among sites (P > 0.05, one-way
ANOVA; depths pooled; Table 3). Similarly, neither rate of ramet production per
colony, cumulative crevice length, nor area of partial colony mortality varied signifi-
cantly between sites and depths (P > 0.05, one- and two-way ANOVAs); nor were any
significant higher-order interactions detected.
REPRODUCTIVE SIZE.—Fission occurred in colonies of all sizes (5 cm2–2.17 m2)
and in colonies across the full range of tissue thicknesses (0.5 mm–6.0 cm high; Fig.
2). The minimum size of colonies observed reproducing by edge fission, polyp-clus-
ter release, and tissue outgrowth was 5.0, 97.0, and 89.0 cm2, respectively. A highly
significant positive linear relationship existed between colony area and number of
ramets produced per colony (Fig. 7; log10 y = 0.3281 * log10 x - 0.5869, P < 0.001; n =
311; 95% conf. limits for slope = 0.2601–0.3961). The minimum average colony size

Figure 5. Comparison between colonies exhibiting edge fission (n = 247), polyp-cluster release
(n = 137), and tissue outgrowth (n = 58). Colonies utilizing the three respective types of fission
varied significantly with respect to cumulative crevice length per colony, number of polyp-clus-
ters per colony, area of partial mortality per colony, and size of parent colony. Bars represent 95%
confidence intervals.
Table 3A. Comparison of morphological traits in Palythoa caribaeorum with respect to site and depth, including colony size, cumulative crevice length, number
of polyp-clusters per colony, ramet production rate, and partial colony mortality.

Site: PPo PR
Data Shallow Deeper Total Shallow Deeper Total
transform n: 80 81 161 87 72 159
Colony size (cm2) Raw Mean 2,381.3 1,691.1 2,034.10 2,457.30 2,509.70 2,481.00
s.d. 3,317.40 2,256.60 2,846.00 3,493.40 3,431.90 3,454.80
Log10 Mean 2.99 2.93 2.96 2.97 3.06 3.01
s.d. 0.63 0.55 0.59 0.72 0.63 0.68
Cum. crevice length (cm) Raw Mean 29.75 24.8 27.3 32.1 43.9 37.4
s.d. 46.67 33.61 40.50 50.70 53.40 52.10
Log10(Y+1) Mean 1.02 1.02 1.02 1.05 1.23 1.13
s.d. 0.69 0.64 0.67 0.72 0.73 0.73
No. polyp clusters per colony Raw Mean 5.4 7.5 6.4 5.3 8.4 6.7
s.d. 7.70 18.30 14.00 6.70 9.70 8.30
Log10(Y+1) Mean 0.66 0.72 0.69 0.69 0.82 0.75
s.d. 0.30 0.33 0.32 0.26 0.34 0.30
No. ramets produced per colony Raw Mean 4.6 4.9 4.7 3.2 4.4 3.7
s.d. 8.10 11.00 9.90 3.69 4.40 4.00
ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM

Log10 Mean 0.41 0.42 0.42 0.32 0.46 0.39

s.d. 0.39 0.39 0.39 0.36 0.38 0.37
Partial colony mortalilty (cm2) Raw Mean 126.5 96.1 86.5 96.1 65.9 82.4
s.d. 550.70 184.80 397.60 184.80 117.10 158.06
Log10(Y+1) Mean 1.13 1.13 1.13 1.24 1.20 1.22
s.d. 0.91 0.70 0.81 0.92 0.86 0.89
13
 14

Table 3B. ANOVA results indicating no significant variation between sites or depths with respect to rate of ramet production per colony, cumulative crevice length,
or area of partial colony mortality (P > 0.05, one- and two-way ANOVAs); nor were any significant higher-order interactions detected.

Sum of squares Degrees of Mean square (MS) Fs P

(SS) freedom (df)
Mean no. ramets produced per colony
Site 0.023 1 0.023 0.171 n.s
Depth 0.434 1 0.434 3.21 n.s
S×D 0.482 1 0.482 3.57 n.s
Error 31.86 236 0.135

Mean cum. crevice length

Site 0.877 1 0.877 1.914 n.s.
Depth 0.828 1 0.828 1.808 n.s.
S×D 0.199 1 0.199 0.436 n.s.
Error 108.14 236 0.358

Mean no. polyp-groups

Site 0.259 1 0.259 3.245 n.s.
Depth 0.185 1 0.185 2.26 n.s.
S×D 0.115 1 0.115 1.44 n.s.
BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Error 18.84 236 0.0798

Mean area of partial colony mortality

Site 0.0002 1 0.0002 2.8x10-4 n.s.
Depth 1.412 1 1.412 2.04 n.s.
S×D 0.16 1 0.16 0.23 n.s.
Error 164.03 236 0.6951
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 15

Figure 6. (A) Frequency of colonies requiring 3, 6, 9, or 12 mo, respectively, for ramet release; and
(B) frequency of ramet release by season at the study sites, varying significantly through time.
Bars represent 95% conf. intervals. n = 320.

that was estimated by the regression model to initiate ramet production (y = 1) was
61.5 cm2.
Ramet production varied significantly among small (< 50 cm diameter), medium
(51–150 cm), and large (> 150 cm) colonies (P < 0.01, one-way ANOVA, n = 320).
This same relationship occurred with respect to mean area of partial mortality. More
specifically, larger colonies exhibited the highest frequencies of fission and partial
mortality (P < 0.05, Tukey-Kramer Test).
Partial mortality accounted for < 5.3% of the area in colonies exhibiting fission
(Table 2). The area of ramets produced and released per colony was significantly larg-
er than the total area lost due to partial coral mortality (P < 0.001, one-way ANOVA;
Table 2). This also held true for ramets produced by edge fission and tissue outgrowth
(P < 0.001, one-way ANOVA), but not for those produced by polyp-cluster release (P
> 0.05, one-way ANOVA).
RAMET SIZE.—Ramet size did not vary significantly among the different fission
variants (P > 0.05, one-way ANOVA; Table 1). Eighty percent of the ramets were be-
tween 0.25 and 36 cm2, irrespective of fission type. The largest ramet observed was
1518 cm2 in size and was produced by polyp-cluster release (Table 2).

DISCUSSION

Duerden documented the process of fission in zoanthids as early as 1898. Few pa-
pers describe this process and even fewer describe the different variants that occur in
nature. This is in part due to terminology problems (e.g., fission vs. fragmentation),
the exercise of caution before introducing new but necessary terms, or because the
variants of fission were simply not recognized in the field.
16 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Figure 7. Linear regression between number of ramets produced annually per colony vs colony
area (cm2). Data were transformed by log10 to achieve normalization.

EDGE FISSION.—Edge fission in P. caribaeorum is comparable to asexual reproduc-

tion in several other species, as in Palythoa mammillosa (Duerden, 1898). In Palythoa
howesii Haddon and Shackleton, 1891, colonies divide into two parts, being joined
only at their bases (Calgren, 1937). Yamazato and Isa (1981) described edge fission in
Palythoa tuberculosa Erper, 1791. Tanner (1997; 1999) and Ryland (1997) described
the formation of ramets in Palythoa caesia through frequent lateral constrictions
(shrinkage), resulting in numerous, nodular clumps. Unfortunately, only a few obser-
vations were made in each case, or descriptions were general, making comparisons
difficult. Edge fission, however, appears to be widely employed within Palythoa in the
SW Atlantic (Brazil and the Caribbean) and on the Great Barrier Reef (Calgren, 1937;
Tanner, 1997; Gonzalez and Acosta, 2000).
POLYP-CLUSTER RELEASE.—No analogs of polyp-cluster release within the Cni-
daria could be found in the literature; thus, this represents a first report for the genus
Palythoa. This variant of fission produced the most ramets of the variants described
here (Acosta and Asbarh, 2000). It is not known whether these ramets remain fertile
during or after dispersal/settlement, or whether the larger and/or thicker ramets
have a better chance of survival than ramets produced by other variants (see Karlson,
1988b). The loss of reproductive polyps, however, is most likely compensated for by
gain in cover (for each genet) and an increase in population size.
TISSUE OUTGROWTH.—This is the first report of tissue outgrowth for the genus
Palythoa. We found one analog in the British zoanthid Parazoanthus axinellae
(Schmidt, 1862). There, colonies growing on rock faces under substrate-limited con-
ditions produced thread-like, club-tipped stolons, which hung vertically, and were
broken off and dispersed by water movement (P. E. J. Dyrynda, cf. Ryland, 1997).
Larger colonies of the coral Porites also exhibit a related growth response (DeVantier
and Endean, 1989).
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 17

POLYP-BALL PRODUCTION.—This represents a first report for polyp-ball production

in P. caribaeorum and for zoanthids in general. Duerden (1898) previously described
the colony’s surface as uniform, with polyps immersed in a thick coenenchyme, ex-
tending only a few mm above it (also see Haddon and Shackleton, 1891). The surface
of the Brazilian P. caribaeorum colonies was irregular; however, the bumps corre-
sponded to polyp-balls rising above the coenenchyme. Polyp-ball production has
been described to occur in the scleractinian coral Goniopora stokesi Milne-Edwards
and Haime, 1851 (Scheer, 1959; Rosen and Taylor, 1969) under unfavorable envi-
ronmental conditions. In P. caribaeorum, polyp balls were rare and associated only
with substrate irregularity. Winter storm currents facilitated polyp-ball release and
dispersal.
THE ROLE OF FISSION IN DETERMINING POPULATION SIZE.—Fission can affect
population size (Tanner, 1997; 1999) and demography (Cook, 1983; Hughes and Jack-
son, 1985; Shick, 1991) in benthic invertebrates. The total number of ramets pro-
duced here could potentially increase the population size at least two-fold in a short
time, if all survived. The number of ramets and new polyp-clusters produced dur-
ing a single year could potentially increase population size by 225%. We agree with
Cooke’s (1976) conclusions that Palythoa has a high rate of asexual reproduction;
our measured rates were more than three times those reported for P. caesia (Tanner,
1997).
In Brazil, 74.2% of the colonies exhibited fission, similar to the 83% level (n=383)
reported by Gonzalez (1999) for Colombian populations. There, it was estimated that
population size could increase by up to 384% during a summer. These potential rates
of increase are even higher than our Brazilian rates, and may be due to density-de-
pendence (Tanner, 1999). Populations in Colombia are less dense, with lower cover
(19%) than in Brazil (Acosta, 2001). The Colombian populations also had a lower
mean size and therefore may reproduce more by fission (Gonzalez, 1999; also see
Tanner, 1999).
The large number of potential ramets produced, and the diversity of fission vari-
ants producing them, suggest that fission has a substantial influence on fitness and
population size, facilitating colonization of nearby habitats. Population dynamics
and colonization in P. caribaeorum may be more heavily dependent upon asexual
rather than sexual reproduction. Abundance of newly attached ramets produced
asexually was higher in the study plots than that derived from sexually-produced
larvae (Acosta, 1999). McFadden (1991) also found that reproduction is primarily by
asexual ramets in the temperate soft coral Alcyonium sp.; sexual propagules rarely
enter the population. She predicted that sexual reproduction would have a negligible
impact on fitness, and that if asexual reproduction by fission were eliminated en-
tirely, fitness would be greatly reduced, leading to rapid local extinction.
High clone production is associated with high fertility (Acosta and Asbahr, 2000).
This contrasts with reports of infertility in this species in the Caribbean (Fadlallah
et al., 1984; A. Acosta, pers. obs.), and in zoanthids such as P. tuberculosa, where
44%–52% of the polyps can be sterile (Yamazato et al., 1973; also see Ryland, 1997).
A high reproductive effort in P. caribaeorum colonies most likely helps to maintain
high genetic variability, which can enhance survival in this variable environment
characterized by frequent disturbance (e.g., cold fronts; Acosta, 2001). Burnett et al.
(1994, 1995) have confirmed high genetic variability in Zoanthus coppingeri Haddon
18 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

and Shackleton, 1891, populations, which, along with regular environmental distur-
bances, apparently prevent its populations from being dominated by a few clones.
Fission in P. caribaeorum provides a constant supply of ramets, as occurs in the
scleractinian coral Astrangia lajollaensis Durham, 1947 in California (Fadlallah,
1981). Ramet production in P. caribaeorum did not decrease during the summer
when gonads were maturing (see Cooke, 1976; Fadlallah et al., 1984) or later dur-
ing spawning (April and May; Acosta and Asbahr, 2000), which contrasts with the
predictions of Minasian and Mariscal (1979) and Minasian (1982). They state that
gonadal development should be accompanied by a decrease in fission rate. Thus, the
cost of fission may be low.
ENDOGENOUS VS EXOGENOUS CONTROL OF FISSION.—Colony fission may have
a genetic basis, but various factors may also trigger and control that process (also
see Karlson, 1991). In P. caribaeorum, fission appears to be at least in part geneti-
cally programmed, as evidenced by: 1) production of morphological changes (e.g.,
in the coenenchyme); 2) production of polyp-clusters and ramets of specific shapes
and sizes; 3) active severing of basal connecting tissue around a polyp-cluster; 4)
simultaneous detachment of a polyp-cluster from its neighbors and the substratum;
and 5) tissue growing into a specific shape to facilitate breakage or colonization of
nearby habitats. By comparison, in the Octocorallia, autotomy of growing tips and a
high level of ramet production in large colonies also appears to be under endogenous
control (Lasker, 1983; Dinesen, 1985), as seen in the gorgonian Junceella fragilis Rid-
ley, 1884 (Walker and Bull, 1983) and the alcyonacean Dendronephthya hemprichi
Klunzinger, 1877 (Dahan and Benayahu, 1997). In the Cnidaria and other inverte-
brates (e.g., asteroids), fission, tissue resorption, and stolon degeneration are also
considered to be under endogenous control (Muirhead and Ryland, 1985; Karlson,
1986; Hughes, 1989).
Abiotic or biotic exogenous factors (e.g., substratum limitation, low light levels,
competition for space, high colony density) also control fission in sea anemones, as-
teroids, and polychaetes (Cook, 1985; McFadden, 1986; Shick, 1991; Mladenov, 1996;
Tanner, 1999) and may help trigger the particular variant of fission employed by P.
caribaeorum and the quantity of ramets produced (see Schmidt, 1970; Minasian,
1976). Fission in P. caribaeorum may be a response to constraints on colony growth
(see Cameron and Endean, 1985; Devantier and Endean, 1989), such as substrate limi-
tation. Edge fission may allow colonies to escape competition for space, an important
potential selective force (see Sebens, 1982b; Oliver, 1984; Jackson et al., 1985). Edge
fission enhances population size and may increase genet survivorship by spreading
risk of mortality over a larger number of ramets (see Cook, 1978; Stoner, 1989). Fit-
ness may be better enhanced by fission rather than by simple colony growth, particu-
larly where substrata are limiting (see Sebens, 1979; Stoner, 1989).
Polyp-cluster release may also represent a strategy to overcome substrate-limita-
tion (see Stoner, 1989). Once a colony monopolizes all available horizontal substra-
tum, polyps begin to grow up vertically —sometimes up to 6 cm; normally, they only
reach 1–3 cm in height (Duerden, 1898; Gleibs, 1994; A. Acosta pers. obs.). Once
space has been monopolized, polyp-cluster release provides space for the colony by
removing tissue and simultaneously promoting dispersal. In addition, the colony
continues to re-colonize vacated areas. Space made available through ramet release
is easily colonized by competitors and is usually recovered by the colony over time.
Competition for space may be reduced via the production of palytoxin (see Gleibs,
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 19

1994), allowing only a few resistant species (e.g., Zoanthus) to colonize, and allowing
P. caribaeorum to colonize and maintain space for long periods of time.
Colonies at > 4 m depth exhibited a high number of polyp-clusters and high fission
rates (A. Acosta pers. obs.). The lower light levels (Acosta, 2001) may limit this spe-
cies’ vertical distribution (see Suchanek and Green, 1981; Sebens, 1982a). Palythoa
colonies may have difficulty sustaining positive energy production in slightly deeper
(3.4 m) vs. shallower (1.4 m) water (Tun et al., 1997). Our Brazilian populations may
experience starvation, as evidenced by their entry into diapause (a long period of
dormancy, characterized by a mucus coating over the colony; Acosta et al., unpubl.
data) during the winter. Starvation and low light levels are known to favor clone
formation (Minasian, 1976; Smith and Lenhoff, 1976; Sebens, 1979). DeVantier and
Endean (1989) have also suggested that low light levels in deeper waters may increase
fission rates in Porites. Ryland (1997) points out that asexual reproduction helps
maintain small colony size in Isaurus, Protopalythoa, Zoanthus, and Palythoa cae-
sia. A small colony size may decrease the total cost of colony maintenance (see Shick,
1976; Bucklin, 1987) in Palythoa caribaeorum by keeping the colonies at an optimal
size for the deeper water environment. At this point, we do not know whether the
decrease in colony size resulting from edge fission is a response to limited energy
input (Sebens, 1980; Minasian, 1982; Steen and Muscatine, 1984), low food availabil-
ity (Okamura, 1985; McFadden, 1986; Anthony, 1997), or maintenance of constant
ratios of circumference to total area (also noted in Palythoa tuberculosa Esper, 1791;
Yamazato and Isa, 1981).
Intraspecific competition for space by densely packed colonies in shallower water
depth suppresses fission in this species. This agrees with Tanner’s (1999) conclusion
that P. caesia fission is density-dependent and decreases at higher densities. Colo-
ny density also has a negative impact on fission and colony growth in scleractinian
corals (Fadlallah, 1982), soft corals (Karlson et al., 1996), and fissiparous asteroids
(Mladenov, 1996).
MINIMAL COLONY SIZE REQUIRED FOR FISSION.—Palythoa caribaeorum com-
monly undergoes fission at small colony sizes (5 cm2). Edge fission occurs in Colom-
bian P. caribaeorum colonies as small as 4 cm2 in area (Gonzalez, 1999); similarly, P.
caesia undergoes fission at <10 cm2 (Tanner, 1997; also see Haddon and Shackleton,
1891). This may be selectively advantageous because it helps to increase metabolic
rate, growth, and reproduction (Ryland and Warner, 1986; Stoner, 1989), genet size
(Chia, 1976; Coates and Jackson, 1985), and efficiency of food capture (McFadden,
1986).
Palythoa caribaeorum colonies as small as 7 cm2 and ramets as small as 4 cm2
can be sexually mature (Acosta and Asbahr, 2000), being similar to observations by
Kimura et al. (1972) and Yamazato et al. (1973) on small colonies of P. tuberculosa.
Reproductive output in P. caribaeorum was positively correlated with colony size
(also see Gonzalez, 1999; Acosta and Asbahr, 2000), suggesting that fitness may be
enhanced by increasing colony size, as is known to occur in P. caesia (Tanner, 1997),
Alcyonium sp. (McFadden, 1991), and some asteroids (Ottesen and Lucas, 1982). The
employment of both sexual and asexual reproductive strategies, even at small colony
sizes, may help to explain P. caribaeorum’s dominance in shallow benthic areas.
SIZE SELECTION AND SURVIVAL OF RAMETS.—Ramet survivorship is a function of
size in many plants and clonal organisms (Highsmith et al., 1980; Newell et al., 1981;
Dahan and Benayahu, 1997). If ramet survival is size-dependent in P. caribaeorum,
20 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

then high mortality may be expected in smaller size classes, and fission may have
less of an impact on population size than would otherwise be expected. On the other
hand, active ramet production over a wide size spectrum (>36–1518 cm2) may help
to compensate for size-dependent mortality. The observed low recruitment of ramets
in P. caribaeorum (Acosta, 1999) may be explained by high ramet mortality or high
dispersal capabilities.
A “trade off” may exist between ramet size, distance of dispersal, and survival,
as Wulff (1985) has suggested for sponges. Clearly, smaller ramets can be carried
over longer distances by local currents. Larger ramets may be negatively buoyant, but
their survival time might be longer due to a larger feeding area or greater lipid re-
serves as in Z. solanderi (see Karlson, 1986, 1988b). Different sized ramets will allow
P. caribaeorum to disperse polyp groups over a wide range of distances, depending
upon local current velocity and direction.
Ramet reproduction was not associated with seasonal temperature fluctuations
(15–29ºC). In anemones and holothurians, temperature can increase fission rates
(Johnson and Shick, 1977; Minasian, 1979; Emson and Mladenov, 1987; Chao and
Tsai, 1995). Palythoa caribaeorum even produced ramets during the winter, which
was characterized by low temperatures (15º–17ºC), storm waves, and high currents,
turbidity, and sedimentation (see Acosta, 2001). Sedimentation rate was an order of
magnitude higher in winter (124.5 ± 145 mg cm-2 d-1) than in the other seasons (con-
sidered sub-lethal for scleractinian corals; Pastorok and Bilyard, 1985; Rogers, 1990).
Perhaps the reason that ramet release was significantly lower in the winter was be-
cause of these more stressful environmental conditions and a higher probability of
ramet mortality. Differences in the degree of stress between sites (sedimentation lev-
els, SSTs, and light) and depths (Acosta, 2001) also did not affect ramet production.
THE RELATIONSHIP BETWEEN PARTIAL MORTALITY AND FISSION.—Some inves-
tigators have suggested that increases in fission rate are a response to partial colony
mortality (Harper, 1977; Cook, 1978; Shick and Hoffmann, 1980; Shostak, 1993; also
see Karlson, 1988a; Shick, 1991); however, we found no evidence to support this. Par-
tial mortality, fission, and fragmentation were positively related to size (Acosta et al.,
2001). Increased frequency of fission was not associated with partial mortality. In
fact, colonies exhibiting edge fission had lower levels of partial colony mortality than
the other variants. In addition, the total area lost due to partial mortality was much
smaller than the total area of ramets produced by fission. Thus, asexual reproduction
in this species does not appear to represent an escape mechanism from partial mor-
tality, as proposed elsewhere (Harper, 1977; Lasker, 1990).
FISSION AS A CONSERVATIVE TRAIT.—Observations of P. caribaeorum in Brazil
(Rio de Janeiro: Angra dos Reis, Ilha Botinas; Recife: Ponta de Galhinas); Colombia
(Santa Marta, González and Acosta, 2000; San Andrés and Providencia Islands, A.
Acosta pers. obs.), Puerto Rico (R. Urueña, Univ. Puerto Rico at Mayagüez, pers.
comm.); and the U.S. (Florida Keys; A. Acosta, pers. obs.) confirm that fission occurs
commonly in the western Atlantic and is a conservative character in this species.
Although populations from the Caribbean and southern Brazil have been geographi-
cally isolated from each other through geological time by the Amazon and Orinoco
rivers (see Rocha and Bowen, 2000), they all exhibit crevice formation, edge fission,
and tissue outgrowth (see González and Acosta, 2000), suggesting that there is a
single western Atlantic species of P. caribaeorum.
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 21

ADVANTAGES OF RAMET PRODUCTION BY FISSION IN PALYTHOA.—Fission has

eight potential selective advantages for Palythoa; 1) Fission may increase fitness by
producing many clones capable of reproduction (McFadden, 1991), thereby increas-
ing the number of gametes representing the parental genotype (Hughes, 1989). It also
decreases risk of mortality in the genets, especially in unpredictable environments
(Cook, 1978); 2) Fission results in an increase in the local dominance of well-adapted
genotypes (Williams, 1975). That is, as long as environmental conditions remain
generally the same, the offspring will share similar levels of survivorship as the par-
ents; 3) Fission permits clones to escape local selective factors, poor-quality micro-
habitats, or environmental stress (Hunter, 1984); 4) It allows the colony to overcome
certain constraints during growth by freeing up space within a colony or through
directional growth (runners); 5) Fission provides a constant supply of ramets, where
sexual propagules are generally produced only seasonally; 6) It increases propagule
survivorship, since ramets are larger than sexually produced larvae (Karlson, 1988b);
7) Fission requires less energy than sexual reproduction (see Fadlallah, 1982; Hunter,
1984; Karlson, 1986). In addition, gamete production and release represent energy
lost from the genet, while ramets represent a direct investment in the genet’s repli-
cates; and 8) Fission regulates population size, clonal structure, and size distribution
in P. caribaeorum (also see Karlson, 1991; McFadden, 1991; Tanner, 1999).
Fission in P. caribaeorum also has several selective disadvantages: 1) It decreases
local genetic diversity in the short-term; 2) reduces the ability of the population to
adapt to a changing environment (Levins 1968; Williams 1975); and 3) results in
lower dispersal capabilities (vs. sexually reproduced larvae).

ACKNOWLEDGEMENTS

Field research was partially supported by the Universidade Estadual de Campinas (UNI-
CAMP, Brazil) - FAEP (#314.97); the Instituto Colombiano para el Desarrollo de la Ciencia
y la Tecnología “Francisco José de Caldas” (COLCIENCIAS, Colombia); Pontificia Universi-
dad Javeriana (Colombia); Louisiana Universities Marine Consortium Foundation (#674042);
Louisiana Universities Marine Consortium (LUMCON, Chauvin); and Louisiana State Uni-
versity, Baton Rouge, Louisiana (USA). We thank the Centro de Biologia Marinha (CEBIMar),
Universidade de São Paulo, Brazil, for supplying research space and equipment for the project.
P. M. Yoshioka and several anonymous reviewers provided helpful comments on the manu-
script.

LITERATURE CITED

Acosta, A. 1999. Ecologia da reprodução assexuada de Palythoa caribaeorum (Zoanthidea:

Cnidaria). Tese de Doutor, Instituto de Biologia, Universidade Estadual de Campinas, S.P.
Brazil. 215 p.
________. 2001. Disease in zoanthids: dynamics in space and time. Hydrobiologia 460: 113–
130.
________ and M. Asbahr. 2000. Reproductive effort in Palythoa caribaeorum. Proc. 9th Int.
Coral Reef Symp., Bali. Abstract, 295.
________, P. W. Sammarco, and L. F. Duarte. 2001. Asexual reproduction in a zoanthid by frag-
mentation. Bull. Mar. Sci. 68: 363–381.
Akcakaya, H. R., M. A. Burgman, and L. R. Ginzburg. 1997. Applied population ecology: Prin-
ciples and computer exercises using RAMAS Ecolab 1.0. Applied Biomathematics, Se-
tauket. 255 p.
22 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Anthony, K. R. N. 1997. Prey capture by the sea anemone Metridium senile: effects of body size,
flow regime, and upstream neighbors. Biol. Bull. 192: 73–86.
Ayre, D. J. and J. M. Resing. 1986. Sexual and asexual production of planulae in reef corals. Mar.
Biol. 90: 187–190.
Bell, G. 1982. The masterpiece of nature: the evolution and genetics of sexuality. Croom Helm,
London and Canberra. 635 p.
Burnett, W. J, J. A. H. Benzie, J. A. Beardmore, and J. S. Ryland. 1994. High genetic variability
and patchiness in a common Great Barrier Reef zoanthid (Palythoa caesia). Mar. Biol. 121:
153–160.
___________, _____________, _____________, and __________. 1995. Patterns of genetic sub-
division in populations of a clonal cnidarian, Zoanthus coppingeri, from the Great Barrier
Reef. Mar. Biol. 122: 665–673.
Bucklin, A. 1987. Growth and asexual reproduction of the sea anemone Metridium: compara-
tive laboratory studies of three species. J. Exp. Mar. Biol. Ecol. 110: 41–52.
Cairns, S. D. 1988. Asexual reproduction in solitary scleractinia. Proc. 6th Int. Coral Reef Symp.,
Townsville. 2: 641–645.
Calgren, O. 1937. Ceriantharia and Zoantharia. British Museum (Natural History), Great Bar-
rier Reef Expedition 1928–1929, Scientific Reports, vol. V, no. 5: 25–209.
Cameron, A. M. and R. Endean. 1985. Do long-lived species structure coral reefs ecosystems.
Proc. 5th Int. Coral Reef Congr., Tahiti. 6: 211–215.
Chao, S. M. and C. C. Tsai. 1995. Reproduction and population dynamics of the fissiparous
brittle star Ophiactis savignyi (Echinodermata: Ophiuroidea). Mar. Biol. 124: 77–83.
Chia, F. S. 1976. Sea anemone reproduction: Patterns and adaptive radiations. Pages 261–269
in G. O. Mackie, ed. Coelenterate ecology and behavior. Plenum Press, New York.
Coates, A. G. and J. B. C. Jackson. 1985. Morphological themes in the evolution of clonal and
aclonal marine invertebrates. Pages 67–106 in J.B.C. Jackson, L.W. Buss, and R.E. Cook,
eds. Population biology and evolution of clonal organisms. Yale University Press, New Ha-
ven.
Cook, R. E. 1978. Asexual reproduction: A further consideration. Amer. Nat. 113: 769–772.
_________. 1983. Clonal plant populations. Am. Sci. 71: 244–253.
_________. 1985. Growth and development in clonal plant populations. Pages 259–296 in
J.B.C. Jackson, L.W. Buss, and R.E. Cook, eds. Population biology and evolution of clonal
organisms. Yale University Press, New Haven.
Cooke, W. J. 1976. Reproduction, growth, and some tolerances of Zoanthus pacificus and Pa-
lythoa vestitus in Kaneohe Bay, Hawaii. Pages 281–288 in G.O. Mackie, ed. Coelenterate
ecology and behavior. Plenum Press, New York.
Crowell, S. and S. Oates. 1980. Metamorphosis and reproduction by transverse fission in an
Edwardsiid anemone. Pages 139–142 in P. Tardent and R. Tarden, eds. Developmental and
cellular biology of coelenterates, Elsevier Press, Amsterdam.
Dahan, M. and Y. Benayahu. 1997. Clonal propagation by the azooxanthellate octocoral Den-
dronephthya hemprichi. Coral Reefs 16: 5–12.
DeVantier, L. M. and R. Endean. 1989. Observations of colony fission following ledge-forma-
tion in massive reef corals of the genus Porites. Mar. Ecol. Prog. Ser. 58: 191–195.
Diaz, J. M., L. M. Barrios, M. H. Cedales, J. Garzon-Ferreira, J. Geister, M. Lopez-Victoria, G. H.
Ospina, F. Parra-Velandia, J. Pinzon, B. Vargas-Angel, F. A. Zapata, and S. Zea. 2000. Áreas
coralinas de Colombia. Serie Publicaciones Especiales, No. 5. INVEMAR, Santa Marta.
176 p.
Dinesen, Z. D. 1985. Aspects of the life history of a stolon-bearing species of Efflatounaria
(Octocorallia: Xeniidae). Proc. 5th Int. Coral Reef Congress, Tahiti 6: 89–94.
Duerden, J. E. 1898. Jamaican Actiniaria, Part I. The Zoantheae. Trans. Roy. Dub. Soc., N.S. 6:
329–384.
Emson, R. H. and P. V. Mladenov. 1987. Studies of the fissiparous holothurian Holothuria par-
vula (Echinodermata: Holothuroidea). J. Exp. Mar. Biol. Ecol. 111: 195–211.
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 23

Fadlallah, Y. H. 1981. The reproductive biology of three species of corals from central Califor-
nia. Ph.D. Diss., University of California, Santa Cruz. 213 p.
________. 1982. Reproductive ecology of the coral Astrangia lajollaensis: Sexual and asexual
patterns in a kelp forest habitat. Oecologia 55: 379–388.
________, R. H. Karlson, and K. P. Sebens. 1984. A comparative study of sexual reproduction
in three species of Panamanian zoanthids (Coelenterata: Anthozoa). Bull. Mar. Sci. 35:
80–89.
Gleibs, S. 1994. Role of a toxic secondary metabolite in the ecology of the genus Palythoa
(Anthozoa) a zoanthid of the Caribbean. Institut für Allgemeine und Spezielle Zoologie,
Biologie, Thesis, Justus-Liebig-Universität Gießen, Germany. 107 p.
González, A. M. 1999. Efecto del tamaño colonial en la frecuencia de fisión de Palythoa carib-
aeorum (Cnidaria: Zoanthidea) en Santa Marta, Caribe Colombiano. Pontificia Universidad
Javeriana, Facultad de Ciencias Básicas, Biología, Bogotá, Tesis de Grado de Biólogo. 60 p.
_____________ and A. Acosta. 2000. Colony size, fission, and effect of ramets to population
growth in the zoanthid Palythoa caribaeorum. Page 308 in 9th Int. Coral Reef Symp., Bali,
Indonesia.
Haddon, A. C. and A. M. Shackleton. 1891. Actiniae: I Zoantheae. Reports on the Zoological
Collections made in Torres Straits by A.C. Haddon, 1888-1889. Scient. Trans. R. Soc. Ser.
2, 4: 674–699.
Harper, J. L. 1977. Population biology of plants. Academic Press, New York. 890 p.
Harrison, P. L. and C. C. Wallace. 1990. Coral reproductive biology. Pages 181–183 in Z. Du-
binsky, ed. Ecosystems of the world, 25, coral reefs. Elsevier, New York.
Highsmith, R. C. 1982. Reproduction by fragmentation in corals. Mar. Ecol. Prog. Ser. 7: 207–
226.
_____________, A. C. Riggs, and C. M. DÁntonio. 1980. Survival of hurricane-generated coral
fragments and a disturbance variant of reef calcification growth rates. Oecologia 46: 322–
329.
Hughes, R. N. 1989. A functional biology of clonal animals. Chapman and Hall, New York.
331 p.
Hughes, T. P. and J. B. C. Jackson. 1980. Do corals lie about their age? Some demographic con-
sequences of partial mortality, fission and fusion. Science 209: 713–715.
___________ and J. B. C. Jackson. 1985. Population dynamics and life histories of foliaceous
corals. Ecol. Monogr. 55: 141–166.
Hunter, T. 1984. The energetics of asexual reproduction: Pedal laceration in the symbiotic sea
anemone Aiptasia pulchella. J. Exp. Mar. Ecol. 83: 127–147.
Jackson, J. B. C., L. W. Buss, and R. E. Cook. 1985. Population biology and evolution of clonal
organisms. Yale University Press, New Haven. 465 p.
Johnson, L. L. and J. M. Shick. 1977. Effects of fluctuating temperature and immersion on
asexual reproduction in the intertidal sea anemone Haliplanella luciae (Verrill) in labora-
tory culture. J. Exp. Mar. Biol. Ecol. 28: 141–149.
Karlson, R. H. 1983. Disturbance and monopolization of a spatial resource by Zoanthus socia-
tus (Coelenterata, Anthozoa). Bull. Mar. Sci. 33: 118–131.
____________. 1986. Disturbance, colonial fragmentation, and size-dependent life history vari-
ation in two coral reef cnidarians. Mar. Ecol. Prog. Ser. 28: 245–249.
____________. 1988a. Size-dependent growth in two zoanthid species: A contrast in clonal
strategies. Ecology 69: 1219–1232.
____________. 1988b. Growth and survivorship of clonal fragments in Zoanthus solanderi
Lesueur. J. Exp. Mar. Biol. Ecol. 123: 31–39.
____________. 1991. Fission and the dynamics of genets and ramets in clonal cnidarian popula-
tions. Hydrobiologia 216: 235–240.
____________, T. P. Hughes, and S. R. Karlson. 1996. Density-dependent dynamics of soft coral
aggregations: the significance of clonal growth and form. Ecology 77: 1592–1599.
24 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Kimura, S., Y. Hashimoto, and K. Yamazato. 1972. Toxicity of the zoanthid Palythoa tubercu-
losa. Toxicon. 10: 611–617.
Kramarsky, E. W., M. Fine, and Y. Loya. 1997. Coral polyp expulsion. Nature 387: 137.
Larson, K. S. and R. J. Larson. 1982. On the ecology of Isaurus duchassaingi (Cnidaria: Zoan-
thidea) from South Water Cay, Belize. Pages 475–488 in K. Rutzler and I.G. Macintyre, eds.
The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay Belize, Smithsonian Contrib. Mar.
Sci, Washington D.C.
Lasker, H. R. 1983. Vegetative reproduction in the octocoral Briareum asbestinum. J. Exp. Mar.
Ecol. 72: 157–169.
__________. 1990. Clonal propagation and population dynamics of a gorgonian coral. Ecology
71: 1578–1589.
Levins, R. 1968. Evolution in changing environments. Princeton University Press, Princeton.
120 p.
McFadden, C. S. 1986. Colony fission increases particle capture rates of a soft coral: Advan-
tages of being a small colony. J. Exp. Mar. Biol. Ecol. 103: 1–20.
_____________. 1991. A comparative demographic analysis of clonal reproduction in a temper-
ate soft coral. Ecology 72: 1849–1866.
Migotto, A. E. 1997. Anthozoan bleaching on the southeastern coast of Brazil in the summer of
1994. Proc. 6th Conf. on Coelenterate Biology. Amsterdam. 329–335 p.
Minasian, L. L. 1976. Characteristics of asexual reproduction in the sea anemone Haliplanella
luciae reared in the laboratory. Pages 289–298 in G.O. Mackie, ed. Coelenterate ecology
and behavior. Plenum Press, New York.
____________. 1979. The effect of exogenous factors on morphology and asexual reproduction
in laboratory cultures of the intertidal sea anemone, Haliplanella luciae (Verrill) (Antho-
zoa: Actiniaria) from Delaware. J. Exp. Mar. Biol. Ecol. 40: 235–246.
____________. 1982. The relationship of size and biomass to fission rate in a clone of the sea
anemone, Haliplanella luciae (Verril). J. Exp. Mar. Biol. Ecol. 58: 151–162.
____________ and R. N. Mariscal. 1979. Characteristics and regulation of fission activity in
clonal cultures of the cosmopolitan sea anemone, Haliplanella luciae. Biol. Bull. 157: 478–
493.
Mladenov, P. V. 1996. Environmental factors influencing asexual reproductive processes in
echinoderms. Oceanol. Acta 19: 227–235.
Muirhead, A. and J. S. Ryland . 1985. A review of the genus Isaurus (Zoanthidea) including new
records from Fiji. J. Nat. Hist. 19: 323–335.
Newell, S. J., O. T. Solbrig, and D. T. Kincaid. 1981. Studies on the population biology of the
genus Viola III The demography of Viola blanda and Viola pallens. J. Ecol. 69: 997–1016.
Okamura, B. 1985. The effects of ambient flow velocity, colony size, and upstream colonies on
the feeding success of Bryozoa. II. Conopeum reticulum, an encrusting species. J. Exp. Mar.
Biol. Ecol. 89: 69–80.
Oliver, J. S. 1984. Selection for asexual reproduction in an Antarctic polychaete worm. Mar.
Ecol. Prog. Ser. 19: 33–38.
Ottesen, P. O. and J. S. Lucas. 1982. Divide or broadcast: Interrelation of asexual and sexual
reproduction in a population of the fissiparous hermaphroditic seastar Nepanthia belcheri
(Asteroidea: Asterinidae). Mar. Biol. 69: 223–233.
Pastorok, R. A. and G. R. Bilyard. 1985. Effects of sewage pollution on coral-reef communities.
Mar. Ecol. Prog. Ser. 21: 175–189.
Rocha, L. A. and B. W. Bowen. 2000. Assessment of cryptic reef fish species in Brazil using
molecular markers. Proc. 9th Int. Coral Reef Symp., Bali. Abstract, 21.
Rogers, C. S. 1990. Responses of coral reefs and reef organisms to sedimentation. Mar. Ecol.
Prog. Ser. 62: 185–202.
Rosen, B. R. and J. D. Taylor. 1969. Reef coral from Aldabra: New mode of reproduction. Sci-
ence 166: 119–120.
 ACOSTA ET AL.: FISSION IN PALYTHOA CARIBAEORUM 25

Ryland, J. S. 1997. Reproduction in Zoanthidea (Anthozoa: Hexacorallia). Invertebr. Reprod.

Dev. 31: 177–188.
___________ and G. F. Warner. 1986. Growth and form in modular animals: ideas on the size
and arrangement of zooids. Phil. Trans. R. Soc. Lond. B. 313: 53–76.
Sammarco, P. W. 1982. Polyp bail-out: An escape response to environmental stress and a new
means of reproduction in corals. Mar. Ecol. Prog. Ser. 10: 57–65.
Scheer, G. 1959. Die formenvielfalt der riffkorallen. Bericht 1958/1959 Naturwissenschaften
Verein Dormstadt 50: 50–67.
Schmidt, H. 1970. Anthopleura stellula (Actiniaria, Actiniidae) and its reproduction by trans-
verse fission. Mar. Biol. 5: 245–255.
Sebens, K. P. 1979. The energetics of asexual reproduction and colony formation in benthic
marine invertebrates. Am. Zool. 19: 683–697.
__________. 1980. The regulation of asexual reproduction and indeterminate body size in the
sea anemone Anthopleura elegantissima (Brandt). Biol. Bull. 158: 370–382.
__________. 1982a. Intertidal distribution of zoanthids on the Caribbean coast of Panama: Ef-
fects of predation and desiccation. Bull. Mar. Sci. 32: 316–335.
__________. 1982b. The limits to indeterminate growth: An optimal size variant applied to pas-
sive suspension feeders. Ecology 63: 209–222.
Shick, J. M. 1976. Ecological physiology and genetics of the colonizing actinian Haliplanella lu-
ciae. Pages 137–146 in G.O. Mackie, ed. Coelenterate ecology and behavior. Plenum Press,
New York.
_________. 1991. A functional biology of sea anemones. Chapman & Hall, New York. 424 p.
_________ and R. J. Hoffmann. 1980. Effects of the trophic and physical environments on asex-
ual reproduction and body size in the sea anemone Metridium senile. Pages 211–216 in P.
Tardent and R. Tardent, eds. Developmental and cellular biology of coelenterates. North-
Holland Biomedical Press, New York.
Shostak, S. 1993. 2. Cnidaria. Pages 45–104 in K.G. Adiyodi and R.G. Adiyodi, eds. Reproduc-
tive biology of invertebrates. VI. Asexual propagation and reproductive strategies. John
Wiley & Sons, New York.
Smith, N. and M. Lenhoff. 1976. Regulation of frequency of pedal laceration in a sea anemone.
Pages 117–125 in G.O. Mackie, ed. Coelenterate ecology and behavior. Plenum Press, New
York.
Soong, K., Y. S. Shiau, and C. P. Chen. 1995. Morphological and reproductive variations of
the zoanthid, Sphenopus marsupialis, from two locations in Taiwan. Proc. 6th Int. Conf.
Coelenterate Biol., Amsterdam. 91 p.
_______, __________, and ________. 1999. Morphological and life history divergence of the
zoanthid, Sphenopus marsupialis off the Taiwanese coast. Zool. Stud. 38: 333–343.
Sokal, R. R. and F. J. Rohlf. 1995. Biometry. W.H. Freeman and Co., New York. 887 p.
Steen, R.G. and L. Muscatine. 1984. Daily budgets of photosynthetically fixed carbon in symbi-
otic zoanthids. Biol. Bull. 167: 477–487.
Stoner, D. S. 1989. Fragmentation: a mechanism for the stimulation of genet growth rates in an
encrusting colonial ascidian. Bull. Mar. Sci. 45: 277–287.
Suchanek, T. H. and D. J. Green. 1981. Interspecific competition between Palythoa caribaeo-
rum and other sessile invertebrates on St.Croix reefs, US Virgin Islands. Proc. 4th Int. Coral
Reef Symp., Manila 2: 679–684.
Tanner, J. E. 1997. The effects of density on the zoanthid Palythoa caesia. J. Anim. Ecol. 66:
793–810.
__________. 1999. Density-dependent population dynamics in clonal organisms: a modeling
approach. J. Anim. Ecol. 68: 390–399.
Tun, K. P. P, A. C. Cheshire, and L. M. Chou. 1997. Twenty-four hour in situ monitoring of
oxygen production and respiration of the colonial zoanthid Palythoa. Environ. Monit. As-
sess. 44: 33–43.
26 BULLETIN OF MARINE SCIENCE, VOL. 76, NO. 1, 2005

Vaughan, T. W. and J. W. Wells. 1943. Revision of the suborders families, and genera of the
Scleractinia. Geol. Soc. of America, Baltimore. 363 p.
Verrill, A. E. 1869. Notes in radiata. Review of the corals and polyps of the west coast of Amer-
ica. Trans. Connect. Acad. I: 1868–1870.
Walker, T. A. and G. D. Bull. 1983. A newly discovered method of reproduction in gorgonian
coral. Mar. Ecol. Prog. Ser. 12: 137–143.
Williams, G. C. 1975. Sex and evolution. Princeton University Press, Princeton. 210 p.
Wulff, J. L. 1985. Dispersal and survival of fragments of coral reef sponges. Proc. 5th Int. Coral
Reef Congr., Tahiti 5: 119–124.
Yamazato, K., F. Yoshimoto, and N. Yoshihara. 1973. Reproductive cycle of a Zoanthid, Palythoa
tuberculosa (Esper). Publ. Seto. Mar. Biol. Lab. 20: 275–283.
___________ and T. Isa. 1981. Dynamics of colonial growth in a zoanthid, Palythoa tuberculosa
(Esper). Proc. 4th Int. Coral Reef Symp., Manila 2: 760 p.

DATE SUBMITTED: 23 November, 2001. DATE ACCEPTED: 29 April, 2004.

ADDRESSES: (A.A) Departamento de Biología, Pontificia Universidad Javeriana, Carrera 7

#43-82, Edificio 53, Oficina 106, Bogotá, Colombia; Depto. Zoologia, IB. UNICAMP, Caixa
Postal 6109, CEP 13083-970 Campinas, S.P. Brazil; Louisiana Universities Marine Consor-
tium (LUMCON), 8124 Highway 56, Chauvin, Louisiana 70344. (P.W.S.) Louisiana Universi-
ties Marine Consortium (LUMCON), 8124 Highway 56, Chauvin, Louisiana 70344. (L.F.D.)
Depto. Zoologia, IB. UNICAMP, Caixa postal 6109, CEP 13083-970 Campinas, S.P. Brazil;
Centro de Biologia Marinha (CEBIMar), Universidade de São Paulo, São Paulo, Brazil. COR-
RESPONDING AUTHOR: (A.A.)