H.M. Guzma?n ? C.A. Guevara ? I.C. Herna?ndez
Reproductive cycle of two commercial species of sea cucumber
(Echinodermata: Holothuroidea) from Caribbean Panama
Received: 28 February 2002 /Accepted: 15 August 2002 / Published online: 18 October 2002

 Springer-Verlag 2002
Abstract The reproductive status of the holothuroid
species Isostichopus badionotus (Selenka, 1867) and
Holothuria mexicana (Ludwig, 1875) was studied over
16 months in Bocas del Toro (Panama), from Novem-
ber 1999 to February 2001. Sexual reproduction was
evaluated by the gonad index method, and by histology
of gonad development. In addition, population struc-
ture was assessed based on sex ratio, minimum repro-
ductive size, and length and weight distributions of
males and females. The sex ratio in both species was 1:1,
with a unimodal population distribution composed
mainly of mature individuals. The minimum reproduc-
tive length and weight were 13?20 cm and 150 g, re-
spectively, for both species, although reproductive
individuals 10 cm in length were also found. A consis-
tently higher gonad index was observed in H. mexicana,
due to a high proportion of mature females and males
and high gonad indices in most monthly samples. Ga-
metogenesis and spawning patterns seemed to occur
throughout the year, with periods of enhanced activity.
Two periods of maximum reproductive activity were
tentatively identi?ed: July?November for I. badionotus
and February?July for H. mexicana, but neither species
had a single, sharply de?ned annual spawning event.
Further work on these exploited holothuroids should
examine the relationships between reproduction and
environmental factors and between reproductive status
and recruitment.
Introduction
During the two last decades, numerous holothuroid
species have been ?shed to critical levels, causing the
imminent collapse of local and regional populations
(Sloan 1985; Richards et al. 1994; Conand 1990, 1997;
Herrero-Pe?rezrul et al. 1999). The demand for Be?che-
de-mer in the Asian market, which has led to over?sh-
ing, has underlined the need for basic studies on the
natural history of holothuroids so that the resource can
be managed wisely. In particular, the reproductive bi-
ology of species of commercial importance has been a
priority in di?erent areas of the world, especially for the
families Stichopodidae and Holothuriidae (Conand
1993a; Chao et al. 1994; Reichenbach et al. 1996; Her-
rero-Pe?rezrul et al. 1999; Ramofa?a et al. 2000). Other
studies have focused on the extrinsic factors in?uencing
the reproductive cycle of the Holothuroidea, such as
temperature (Ong-Che 1990; Sewell 1992; Hopper et al.
1998), precipitation (Kubota and Tomari 1998), salinity
(Ong-Che 1990), monsoon runo? (Krishnaswamy and
Krishnan 1967), lunar phase (Kubota and Tomari 1998),
and active photosynthetic radiation (Catalan and
Yamamoto 1994).
The information available on this fundamental pro-
cess for commercially important Caribbean holothu-
roids, however, is mostly limited to subtropical areas
(Engstrom 1980; Mosher 1982; van Veghel 1993; Pires-
Nogueira et al. 2001), and more precise studies on
reproduction have been recommended (Buitrago and
Boada 1996; de la Fuente-Betancourt et al. 2001). The
highest gonad index for Isostichopus badionotus was
observed during late spring (October?November) and
summer (January?February) months in subtropical
Atlantic Brazil, and spawning was reported only in
January when water temperature was maximal (Pires-
Nogueira et al. 2001). Spawning in I. badionotus was
observed after the full moon of August in tropical Bo-
naire, Netherlands Antilles (Graaf et al. 1999). Similarly,
gametogenesis and spawning of Holothuria mexicana
Marine Biology (2003) 142: 271?279
DOI 10.1007/s00227-002-0939-x
Communicated by J.P. Grassle, New Brunswick
H.M. Guzma?n (&) ? C.A. Guevara ? I.C. Herna?ndez
Smithsonian Tropical Research Institute,
Apartado Postal 2072, Balboa, Panama
Mailing address: H.M. Guzma?n
Smithsonian Tropical Research Institute,
Unit 0948, APO AA, 34002-0948, USA
e-mail: guzmanh@naos.si.edu
Fax: +1-507-2128790
from southern Florida occurred during spring?summer
and late summer, respectively (Engstrom 1980; Mosher
1982). Resorption of relict gametes followed, but the
breeding season was considered unclear due to the
presence of enlarged gonads full of mature gametes from
September?November (Engstrom 1980). Van Veghel
(1993) reported spawning for >70% of H. mexicana
individuals within the ?rst 5 days following the full
moons between August and October on Curac?ao reefs,
perhaps the only information available for tropical ar-
eas. The minimum length at maturity for I. badionotus is
reported to be 18 cm (Rodr??guez-Milliet and Pauls
1998), while lengths of 57.9 and 75.7 cm have been re-
ported for H. mexicana males and females, respectively,
from Florida (Engstrom 1980).
Non-traditional ?shing areas in the Caribbean have
been targeted for supplying the Asian market in recent
years (Buitrago and Boada 1996; Rodr??guez-Milliet and
Pauls 1998; de la Fuente-Betancourt et al. 2001), with
species of commercial importance. In the present study,
we describe aspects of the reproductive biology of two of
these species, I. badionotus and H. mexicana. Both spe-
cies are widely distributed throughout the Caribbean.
They commonly inhabit shallow waters with sandy,
coral, or interspersed seagrass areas, and prefer seagrass
meadows (Humann 1992; Hendler et al. 1995). Both
species can reach lengths of up to 50 cm (Hendler et al.
1995). Both species, like most of the Holothuroidea, are
gonochoric.
The goals of the study were: (1) to describe the annual
reproductive cycle, using the gonad index (GI) method,
con?rmed by histological examination of gonads; (2) to
compare the annual reproductive cycle data with envi-
ronmental variables (rainfall, temperature); and (3) to
analyze the population structure (size frequency by sex,
sex ratio, minimum reproductive size) in the Archipelago
of Bocas del Toro, Panama.
Materials and methods
Sexual reproduction assessment
Specimens of Isostichopus badionotus (Selenka, 1867) and of
Holothuria mexicana (Ludwig, 1875) were collected at Punta
Pondsack (0917?19??N; 08219?43??W) and at Punta Quarys
(0916?10??N; 08223?18??W), in the archipelago of Bocas del Toro,
Panama. Adults and juveniles of both species were commonly
found in sandy areas with abundant seagrass (Thalassia testudi-
num), at depths of up to 5 m. The geology, climate, oceanogra-
phy, and orography of the archipelago have been described
previously (Rodr??guez et al. 1993; Greb et al. 1996; Guzma?n and
Guevara 1998). About 30 mature individuals of each species were
collected monthly just before or on the day of full moon, from
November 1999 to February 2001 (475 and 479 individuals per
species, respectively). We expected gonad weight to be maximal at
the time of full moon according to van Veghel (1993) and Graaf
et al. (1999), who observed spawning after the full moon in
Curac?ao and Bonaire, respectively. These holothurians were
placed in buckets of 
10 l of seawater and ca. 5?10 g of MgSO4
for 5?10 min, until completely relaxed and drained, thus facili-
tating biometric measurements (sensu Ong-Che 1990; Chao et al.
1993).
Body length was measured for drained individuals (to within
1 mm). For the extraction of gonads, an incision was made through
the ventral body wall, and all internal parts were removed. We
measured weight using gutted individuals (to within 0.01 g), de-
?ned as body wall wet weight (sensu Conand 1981; Chao et al.
1993, 1995). This weight (gutted) was chosen, since drained weight
may include sediments in the digestive tract and water in the re-
spiratory system. We used drained body weight (sensu Conand
1981; Chao et al. 1993) only to obtain the minimum reproductive
size, without sacri?cing more individuals (Ramofa?a et al. 2000).
Body length was compared to gutted weight (Fig. 1). A signi?cant
but weak relationship was observed for both species, although the
r2 was higher in H. mexicana. This may have been due to the
relative rigidity of the body wall in H. mexicana, which allowed
more precise measurement, compared to the smoother, more ?accid
body wall in I. badionotus.
The extracted gonads were weighed separately (0.01 g preci-
sion) to develop a gonad index. Gonad index was calculated as
follows: GI=[(gonad wet weight)/(gutted weight)]?100. Gonads
were then ?xed in Bouin?s solution for 24 h, washed alternately in
water and 50% ethanol several times, and preserved in 70%
Fig. 1 Isostichopus badionotus, Holothuria mexicana. Relationship
between body length and wall wet weight. Correlations were
statistically signi?cant (P<0.001) as were the regressions
(F(1, 473)=104.748 and F(1, 477)=159.735, respectively)
272
ethanol. Histological analysis of gonads was completed following
standard procedures (Humason 1972); gonads were dehydrated in
an ethanol series, embedded in para?n, and sectioned at 8 lm.
Sections were stained in Mayer?s hematoxylin and eosin.
Males and females were identi?ed by gonad morphology and
color (sensu Engstrom 1980; Herrero-Pe?rezrul et al. 1999; Ramo-
fa?a et al. 2000); ovaries were pink, orange or red, with relatively
uniform diameter, while testes were creamy white and nodose
throughout their length. Sex and gonad developmental stage were
con?rmed histologically for each individual. Reproductive indi-
viduals for both species were classi?ed into ?ve stages of sexual
development following the criteria of Ramofa?a et al. (2000): re-
covery (I), growth (II), maturity (III), partly spawned (IV), and
spawned (V). However, individuals in recovery (I) were never
found in our study, due either to the specimens? small size (im-
mature), poor histological quality of the slides, or to the apparent
re-absorption of tubules after spawning (see Sewell and Chia 1994).
The reproductive cycle of the sea cucumbers was compared with
average monthly rainfall, using data obtained from a meteorolog-
ical station at the airport on Isla Colo?n, and with mean monthly
water temperatures. Temperature was measured continuously with
a thermograph at 3 m depth, located south of Isla Colo?n
(0900?57??N; 08215?51??W).
Minimum reproductive size and size distribution
Monthly collections were designed to collect only larger, mature
holothuroids, so those individuals were not used to evaluate min-
imum reproductive size. Size at ?rst reproduction was estimated to
be when 50% of the sampled individuals had recognizable gonads
(sensu Conand 1981). Both species were collected from outside the
monthly collection areas (south of Isla Colo?n) between January
and March 2001, with emphasis on smaller sizes (<25 cm length),
which were assumed to be juveniles. A small incision (<2 cm) was
made in the dorsal region of each specimen to record gonad pres-
ence (Ramofa?a et al. 2000). Length and drained body weight were
measured and the individuals were released. These individuals and
the ones collected monthly were combined only to evaluate popu-
lation sex ratio, and the size distribution of reproductive and non-
reproductive holothuroids.
Statistical analyses
Parametric statistical tests were performed. Data were transformed
appropriately when the variance requirements (homoscedasticity
and normality) were not ful?lled (Sokal and Rohlf 1995). Pearson
product moment correlation and regression analysis were used to
test for relationships among the di?erent biometric parameters
(length, weight, gonad weight). Species di?erences in body size were
tested with a Student?s t-test. To compare the monthly variation in
the gonad index and between sexes, a two-way ANOVA test was
applied, and the di?erences between months were evaluated using
Tukey?s a posteriori test for multiple comparisons.
Results
We observed an overall mean (?SE) length of 27.05?
0.19 cm (range: 16?41 cm) and 32.93?0.25 cm
(13?55 cm), in Isostichopus badionotus and Holothuria
mexicana respectively. Mean gutted weights were
224?2.9 g (82?426 g) and 257?3.7 g (58?566 g),
respectively.
Minimum reproductive size
From the 55 small individuals of I. badionotus haphaz-
ardly collected to determine the minimum reproductive
size, we obtained an overall mean length of 15.5?
0.43 cm, of which 51% were non-reproductive with a
mean length of 13.6?0.48 cm. For the 71 individuals of
H. mexicana, a similar overall mean length of 15.9?
0.46 cm was observed, with 35% being non-reproductive
with a mean length of 13.3?0.59 cm. No di?erences
were found when comparing the overall lengths between
species (t=?0.778, P=0.438). Both species had some
individuals <10 cm in length with mature gonads. Us-
ing the criterion of 50% of ?rst reproductive occurrence,
a length range of 13?15 cm was established as the min-
imum reproductive length for both species (Fig. 2).
Using drained weight of individuals as the criterion
for estimating size at reproductive maturity (gutted
weight was not measured), we obtained non-reproduc-
tive weights of 68.4?4.41 g for I. badionotus and
60.9?5.05 g for H. mexicana. The mean weight of all
individuals was 98.1?7.1 and 97.9?5.4 g, respectively
(55 and 72 individuals, respectively). The observed
minimum weight of reproductive individuals was 50?
100 g in both species; however, 50% of the sea cucum-
bers weighed ca. 150 g in both species (Fig. 2). In spite
of the wide variance observed in the populations, sig-
ni?cant di?erences were not found between the weights
of the two species (t=?0.089, P=0.929).
Sex ratio
Both species were gonochoric and sexes were distin-
guished by color and histology. Of the total number of
sea cucumbers collected in this study, 89.6% of I. bad-
ionotus (n=530) and 92% of H. mexicana (n=551) were
reproductive (with gonads in various stages of develop-
ment). The sex ratios for the two species were similar, 1:1
as expected (v2=0.490, P=0.484 in I. badionotus;
v2=0.434, P=0.510 in H. mexicana).
Weight and size distributions
The size distributions of mature individuals in monthly
samples of both species were evaluated using body wall
weight (gutted) and body length. A Kolmogorov?Smir-
nov normality test indicated normal distributions for
body wall weights in both species (Fig. 3A?D), while
length classes were normally distributed only for males
(Fig. 3F, H). The modal weight for I. badionotus
(n=448) was ca. 200?250 g, unimodal, and somewhat
skewed in females and males (Fig. 3A, B). For H. mex-
icana, the modal weight was ca. 300 g (n=460), also
unimodal, but nearer normal (Fig. 3C, D). When com-
paring the two species based on body length, unimodal
distributions were observed (25?30 cm and 35 cm,
respectively), similar to those for weights (Fig. 3E?H).
Again, the tendency for H. mexicana to be heavier and
longer may be a result of the rigidity of the body.
Both gutted weight and length were very similar
between the sexes in both species. Mean lengths of
273
27.3?0.28 and 27?0.28 cm were observed in females
and males of I. badionotus, respectively, whereas mean
body wall weights were 223.5?3.9 g in females and
230?4.5 g in males. In H. mexicana, mean length was
ca. 33?0.35 cm for both females and males, while mean
weights were ca. 258.0?5.7 and 260?5.2 g in females
and males, respectively.
Changes in gonad developmental stages
and gonad index
Gametogenesis appeared continuous, with periods of
enhanced activity, but there was no clear annual pat-
tern in either species. The mature stage (III) was ob-
served in both sexes and both species, but not
throughout the study (Fig. 4). In I. badionotus, males in
stage II dominated throughout the year, and the pro-
portion in stage III peaked only in November in both
years and in June (Fig. 4A), while females had more
individuals in stages III (mature) and IV (partly
spawned) (Fig. 4B). Over 50% of females were partly
spawned in May, July (100%), and August, and <40%
in January 2000 (Fig. 4B). In H. mexicana both sexes
were mature (stage III) throughout most of the study,
particularly females (Fig. 4C, D). Stage II was never
observed in females of H. mexicana and sporadically in
males, though 100% of males were in stage II in
October and November. The proportion of partly
spawned males (stage IV) peaked in January, August,
and September (100%), while <40% of females were
in stage IV only in November (Fig. 4C, D). Only 40%
of males were observed in stage V (spawning) during
December 1999.
Average gonad index based on body wall weight was
always lower in I. badionotus (2.63?0.24 and 4.73?0.44
in females and males, respectively) than in H. mexicana
(3.2?0.39 and 7.66?0.92 in females and males, re-
spectively). Di?erences between sexes and among
months were signi?cant in both species, though the in-
teraction was not signi?cant for I. badionotus (Table 1).
A Tukey test revealed a similarity between the peaks in
GI observed in March, May?November (both years),
and January 2001 for I. badionotus, while for H. mexi-
cana, July?s peak of maximum activity was di?erent
from all other months, but marginally similar (P=0.054)
to June and February 2001.
Overall mean GI (regardless of sex) in I. badionotus
showed two apparent peaks of reproductive activity, one
very short in March and another more prolonged and
di?cult to de?ne between July and November (Fig. 5A).
Partly spawned females were observed after the ?rst
peak and, unexpectedly, at the beginning of the second
peak (Fig. 4B). GI in females (Fig. 5B) and males
(Fig. 5C) followed the same seasonal pattern as for total
individuals, although a more pronounced reduction in
reproductive activity was observed in females from
November onwards. December?s decline did not
Fig. 2 Isostichopus badionotus,
Holothuria mexicana. Length
and drained wet weight distri-
butions in reproductive and
non-reproductive individuals
(%) used to calculate minimum
reproductive size (only individ-
uals with <25 cm). Values on
top of bars indicate number of
individuals. Sample sizes were
55 and 71, respectively
274
coincide with the presence of partly spawned (IV) or
spawned (V) stages (Fig. 4A, B).
Reproductive activity in H. mexicana seemed more
de?ned. A gradual increase in GI was observed from
February onwards, with a peak during July (Fig. 5).
Partly spawned males (stage IV) were observed imme-
diately after this peak (Fig. 4C). The GI in males was
greater than in females (Fig. 5C). There was a marked
inter-annual di?erence for both species, noticeably be-
tween January?February 2000 and 2001. This last ob-
servation obscured any clearcut seasonal interpretation,
since we could not de?ne whether a peak in reproductive
activity really occurred at the beginning of every year,
though spawning individuals (stage V) were only
observed after that peak (Fig. 4C).
In both species, there may be a relationship between
GIs and sea-surface temperature, but not with rainfall
(Fig. 5D). The two species did not have overlapping
periods of reproductive activity, but the periods seemed
to coincide with warmer temperatures (average
>28.5C) (Fig. 5A). If precipitation in 1999 were
anomalous, it would seem that most reproductive ac-
tivity in I. badionotus occurred during periods of lower
rainfall (yearly average of <276.9 cm), contrary to what
was observed in H. mexicana (Fig. 5D).
Discussion
Reproductive cycle
Isostichopus badionotus and Holothuria mexicana are
very common Holothuroidea throughout the Caribbean,
but little is known about their ecology in tropical areas,
which makes any attempt to compare our data with
other populations in the region di?cult. However, this
study shows that the reproductive behavior of both
species was similar in some respects to most tropical
Holothuroidea; the sexes are separate, gametes are
released, and fertilization is external, and there may be
an annual reproductive cycle with a continuous and
broad peak of reproductive activity when seasonal water
temperatures are higher (Conand 1982, 1993a; Smiley et
al. 1991; Chao et al. 1995; Hendler et al. 1995; Hopper et
al. 1998).
Comparing the limited information available on the
reproductive biology of the two species studied here (GI
only), we found that the timing of maximum activity for
H. mexicana in Bocas del Toro (May?July) occurred up
to 2 months before that in Florida (August?September)
(Engstrom 1980). As in other studies (Engstrom 1980;
Fig. 3A?H Isostichopus bad-
ionotus, Holothuria mexicana.
Frequency distributions (num-
ber of sexed individuals) based
on wall wet weight (left) and
length (right) for female (closed
bars) and male (gray bars) in
I. badionotus (A, B, E, F) and
H. mexicana (C, D, G, H)
275
Hopper et al. 1998; Reichenbach 1999), we observed
numerous sea cucumbers with mature gonad stages
throughout the year, suggesting that individuals may
reproduce continuously or asynchronously several times
during the year. The broad period of enhanced activity
(2?3 consecutive months) as seen for both species, par-
ticularly I. badionotus, has been reported for congeneric
species (Ong-Che 1990; Conand 1993b; Chao et al. 1994,
1995; Hopper et al. 1998; Herrero-Pe?rezrul et al. 1999;
Ramofa?a et al. 2000). This suggests that within broader
annual peaks, multiple spawning events might occur,
rather than one single event per year. This inference is
based on two previous reports on H. mexicana at dif-
ferent latitudes. In Curac?ao, with a latitude and
temperature range similar to Panama and Venezuela,
H. mexicana spawns in September and October (van
Veghel 1993), whereas in Florida (under laboratory
conditions) spawning was induced from July to
September (Mosher 1982).
Generally, temperature is an important factor for
interpreting seasonal variations in the gonad index;
nevertheless, Hopper et al. (1998) have suggested that
??increased duration of elevated sea temperature does
not appear to in?uence the length of the breeding pe-
riod??. For H. mexicana in Florida, the annual tem-
perature range is greater and more markedly seasonal
compared with Panama. In our study, maximal repro-
duction was observed in I. badionotus and H. mexicana
during the warmest part of the year, when the tem-
perature was generally 2C above the annual average
and 3?4C higher than the lowest monthly average
recorded during the study period. Similarly, spawning
of I. badionotus in Brazil occurs in January, when the
temperature reaches its maximum of 30C (Pires-
Nogueira et al. 2001), a relationship widely observed in
other holothuroids (Conand 1981, 1993a; Hopper et al.
1998; Herrero-Pe?rezrul et al. 1999; but see Ramofa?a
et al. 2000). We cannot discount the possibility that, in
the Bocas del Toro archipelago, spawning may be in-
duced by reductions in salinity associated with local
run-o? and rainfall, which regularly a?ect productivity
in coastal ecosystems (see Krishnaswamy and Krishnan
1967; Ong-Che 1990; Kubota and Tomari 1998). Ong-
Che (1990), suggests that, in the tropics, salinity could
be more important than temperature, since the latter is
Fig. 4A?D Isostichopus badionotus, Holothuria mexicana. Percent-
age of individuals with gonads for each of four stages (II?V) by sex,
from November 1999 to February 2001. A, B I. badionotus; C, D H.
mexicana. Samples of <5 individuals not included (sensu Ramo-
fa?a et al. 2000)
Table 1 Isostichopus badiono-
tus, Holothuria mexicana.
Analysis of variance for gonad
index data between sexes and
among months from November
1999 to February 2001
Source df SS MS F P
I. badionotus
Sex 1 2.050 2.050 6.494 0.011
Month 15 42.091 2.806 8.888 <0.001
Sex?Month 15 6.164 0.411 1.302 0.197
Residual 416 131.332 0.316
Total 447 192.245 0.430
H. mexicana
Sex 1 28.971 28.971 38.893 <0.001
Month 15 151.339 10.089 13.545 <0.001
Sex?Month 15 20.735 1.382 1.856 0.026
Residual 428 318.816 0.745
Total 459 533.223 1.162
276
more constant throughout the year. The average sa-
linity in the study area is 30.6& (28.9?35.1&), and
clearly responds to changes in rainfall and run-o? af-
fecting the upper 1-m sea surface layer (Guzma?n and
Guevara 1998). Hence, we considered that these three
parameters (temperature, salinity, rainfall) act together,
but they are di?cult to interpret for this area of Pan-
ama, where it rains practically year-round, with mo-
ments of fast run-o? associated with the orography of
the coastal zone. Furthermore, and less conclusively,
there is considerable inter-annual variation in rainfall
and its distribution within the archipelago (Gordon
1982; Rodr??guez et al. 1993; Guzma?n and Guevara
1998), which could contribute to year-to-year di?er-
ences in spawning patterns. Interestingly, Chao et al.
(1995) pointed out that di?erent methods used to
evaluate spawning periods of holothurians may have
confused the interpretation of factors a?ecting
reproduction.
Maturity stages and gonad index in relation
to size and weight
The gonad index is recognized as a reliable indicator for
the study of sexual reproduction in Holothuroidea
(Chao et al. 1995). Nevertheless, it is possible that the
index is a?ected by small variations in body length,
which are re?ected as signi?cant changes in body weight
(Engstrom 1980). According to Conand (1993b), the
percentage of individuals with mature gonads, and the
gonad index, can be used to estimate reproductive e?ort
or energy invested in reproduction. Furthermore, she
pointed out that body weight could also be important
for interpreting interspeci?c variations, and suggested
the sexes should be analyzed separately. In this study,
there was a notable contrast between the species, with
greater reproductive e?ort observed in H. mexicana,
similar to that in Florida populations (Engstrom 1980).
Among I. badionotus females, 25?100% were in the
mature stage (III) in 9 of the 16 monthly samples, while
for males 35?70% were in this stage in only 2 of
16 samples. This contrasts sharply with H. mexicana,
where >60% of both males and females were mature
throughout the period, with the exception of males in
October and November 2000. Thus, we found that ma-
turity stages and GIs did not clearly explain the annual
cycle.
Similarly, important average di?erences between go-
nad indices of both species were observed; GI for total
individuals was twice as big in H. mexicana as in
I. badionotus (Fig. 5A), very similar between females of
both species (Fig. 5B), and greater in H. mexicana males
than in I. badionotus males (Fig. 5C). Conand (1993b)
suggested that such di?erences may be due to the fact
that individuals (females in her study) consumed more
food, used the absorbed material more e?ciently, or
used it preferentially in the production of gonads. We
suggest that the continuous presence of the mature
individuals for both species and both sexes (except
I. badionotus males), and perhaps the fact that the col-
lections were always made at full moon and apparently
before spawning events, may have resulted in high
gonad weights, hence GIs, during the entire study. Most
studies do not provide details on the lunar phase and
monthly sampling times, so no reliable comparisons can
be made.
First reproductive age and population structure
In our study, as in other holothuroid studies (Conand
1993a; Chao et al. 1994), a unimodal size distribution
was found, composed mostly of reproductive adults.
We observed this for both sexes whether we used body
wall weight or body length in our analyses. When
comparing Panamanian holothuroids with the infor-
mation available for those in Venezuela (Rodr??guez-
Milliet and Pauls 1998), similar lengths were obtained
in I. badionotus (20?30 cm) and much smaller lengths
Fig. 5A?D Isostichopus badionotus, Holothuria mexicana. A?C
Monthly mean and standard errors in gonad index based on body
wall wet weight (dotted line and left axis for I. badionotus; solid line
and right axis for H. mexicana), for all sexed individuals (A),
females (B), and males (C). D Monthly variation in water
temperature at 3 m depth (solid square) and rainfall (solid triangle),
from November 1999 to February 2001, at Isla Colo?n, Bocas del
Toro, Panama
277
in H. mexicana (10?30 cm). The size ranges may not be
entirely representative of populations in Venezuela,
given the small sample size (123 individuals). Con-
cerning sex ratio, we found the expected 1:1 female/
male ratio. This result is similar to those reported for
the majority of Holothuroidea, and speci?cally for
Holothuria and Isostichopus species (Ong-Che 1990;
Conand 1993a; Foster and Hodgson 1995; Herrero-
Pe?rezrul et al. 1999).
A direct relationship between body size and repro-
duction is known for many marine invertebrates (Cata-
lan and Yamamoto 1994). The minimum reproductive
length found here (13?15 cm) was similar for both
species and was about half the modal lengths. It is
important to note that many reproductive individuals
(30?50%) with lengths <15 cm were found in both spe-
cies. Furthermore, we observed reproductive juveniles
with lengths of 10?15 cm, and recorded adults >20 cm
long that lacked gonads completely. The estimated
minimum reproductive length for I. badionotus in Ven-
ezuela was 18 cm (Rodr??guez-Milliet and Pauls 1998),
somewhat larger than that determined in Panama.
Conclusion and management recommendations
We have described the seasonal reproductive pattern of
two commercially important species for Caribbean
Panama. H. mexicana seems to have a greater repro-
ductive e?ort throughout the year compared with
I. badionotus. Although we suspect multiple spawning
events for both species during the year, it is likely that
maximum spawning activity occurs between February
and November. This range is wide when compared to
other species, where the reproductive stage is generally
con?ned to 2?3 months (Conand 1982; Smiley et al.
1991; Chao et al. 1994; Hopper et al. 1998; Herrero-
Pe?rezrul et al. 1999; Ramofa?a et al. 2000). This pattern
might make management decisions easier for these spe-
cies. However, more studies on recruitment patterns and
juvenile survivorship on longer time-scales are recom-
mended to further understand the population ecology of
the two species.
The exploitation of Holothuroidea in Panama, as in
many other Caribbean countries (Buitrago and Boada
1996; Rodr??guez-Milliet and Pauls 1998; de la Fuente-
Betancourt et al. 2001), started with the consent of the
authorities and without any information on the ecology
of the species. The ?shing pressure exerted on this re-
source since 1997 forced the Panamanian government to
close the ?shery in the Bocas del Toro archipelago, al-
though illegal ?shing continues (Cruz 2000). Initially,
the establishment of a partial or total seasonal ban in the
archipelago is recommended; this might include the
suggested period of maximum reproductive activity,
July?November for I. badionotus and February?July for
H. mexicana. Due to the present abundances and ?shing
preferences, we consider that priority should be given to
I. badionotus, and another species not studied here,
which is in a critical state of scarcity, Astichopus mul-
ti?dus (Guzma?n and Guevara 2002).
Acknowledgements This research was partially funded by the
Fundacio?n Natura, the Fundacio?n PROMAR and the Smithsonian
Tropical Research Institute. A. Domingo, A. Gonzalez, L. Par-
tridge and W. Pomaire provided invaluable assistance in the ?eld
and the laboratory. Special thanks to K. Kneebone for language
assistance. We thank the Government of Panama for providing all
necessary permits to work in the country.
References
Buitrago J, Boada JA (1996) La pesca de la holoturia Isostichopus
badionotus en el oriente de Venezuela. Mem Soc Cienc Nat La
Salle 146:33?40
Catalan MAB, Yamamoto M (1994) Annual reproductive cycle of
the Japanese holothurian Eupentacta chronhjelmi. Can J Zool
72:387?396
Chao SM, Chen CP, Alexander PS (1993) Reproductive periodicity
of a tropical dendrochirote holothurian, Phyrella fragilis
(Echinodermata: Holothuroidea), in Taiwan. Bull Inst Zool
Acad Sin (Taipei) 32:111?119
Chao SM, Chen CP, Alexander PS (1994) Reproduction and
growth of Holothuria atra (Echinodermata: Holothuroidea) at
two contrasting sites in southern Taiwan. Mar Biol 119:565?570
Chao SM, Chen CP, Alexander PS (1995) Reproductive cycles of
tropical sea cucumbers (Echinodermata: Holothuroidea) in
southern Taiwan. Mar Biol 122:289?295
Conand C (1981) Sexual reproduction of three commercially im-
portant Holothuria species (Echinodermata) from the lagoon of
New Caledonia. Bull Mar Sci 31:523?543
Conand C (1982) Reproductive cycle and biometric relations in a
population of Actinopyga echinites (Echinodermata: Holothu-
roidea) from the lagoon of New Caledonia, western tropical
Paci?c. In: Lawrence JM (ed) Proc Int Echinoderm Conf
Tampa Bay, vol 1. Balkema, Rotterdam, pp 437?442
Conand C (1990) The ?sheries resources of Paci?c island countries,
part 2. Holothurians. FAO Fish Tech Pap 272:1?143
Conand C (1993a) Ecology and reproductive biology of Stichopus
variegatus an Indo-Paci?c coral reef sea cucumber (Echino-
dermata: Holothuroidea). Bull Mar Sci 52:970?981
Conand C (1993b) Reproductive biology of the holothurians from
the major communities of the New Caledonian lagoon. Mar
Biol 116:439?450
Conand C (1997) Are holothurian ?sheries for export sustainable?
In: Lessios HA, MacIntyre IG (eds) Proc 8th Int Coral Reef
Symp, vol 2. Smithsonian Tropical Research Institute, Balboa,
Panama, pp 2021?2026
Cruz LG (2000) Campan?a de educacio?n ambiental para la con-
servacio?n y proteccio?n del pepino de mar (Echinodermata:
Holothuroidea) en 15 comunidades del Archipie?lago de Bocas
del Toro y Pen??nsula Valiente. Tesis de maestr??a, Universidad
Santa Mar??a La Antigu?a, Panama City
de la Fuente-Betancourt GM, Jesu?s-Navarrete A, Sosa-Cordero E,
Herrero-Perezrul MD (2001) Assessment of the sea cucumber
(Echinodermata: Holothuroidea) as a potential ?shery resource
in Banco Chinchorro, Quintana Roo, Mexico. Bull Mar Sci
68:59?67
Engstrom NA (1980) Reproductive cycles of Holothuria (Halode-
ima) ?oridana, H. (H.) mexicana and their hybrids (Echino-
dermata: Holothuroidea) in southern Florida, USA. Int J
Invertebr Reprod 2:237?244
Foster GG, Hodgson AH (1995) Annual reproductive cycles of
three sympatric species of intertidal holothurians (Echinoder-
mata) from the coast of the eastern cape province of South
Africa. Invertebr Rep Dev 27:49?59
Gordon BL (1982) A Panama forest and shore: natural history and
Amerindian culture in Bocas del Toro. Boxwood, Paci?c
Grove, Calif.
278
Graaf de M, Geertjes GJ, Videler JJ (1999) Observations on
spawning of scleractinian corals and other invertebrates on the
reefs of Bonaire (Netherlands Antilles, Caribbean). Bull Mar
Sci 64:189?194
Greb L, Saric B, Seyfried H, Broszonn T, Brauch S, Gugau G,
Wiltschko W, Leinfelder R (1996) O?kologie und Sedimentolo-
gie eines Rezenten Rampensystemes an der Karibikku?ste von
Panama. Pro?l, Band 10, Universita?t Stuttgart, Stuttgart
Guzma?n HM, Guevara CA (1998) Arrecifes coralinos de Bocas del
Toro, Panama?. I. Distribucio?n, estructura y estado de conser-
vacio?n de los arrecifes continentales de la Laguna de Chiriqu?? y
la Bah??a Almirante. Rev Biol Trop 46:601?622
Guzma?n HM, Guevara CA (2002) Population structure, distribu-
tion and abundance of three commercial species of sea cu-
cumber (Echinodermata) in Panama. Caribb J Sci (in press)
Hendler G, Miller JE, Pawson DL, Kier PM (1995) Echinoderms
of Florida and the Caribbean sea stars: sea urchins and allies.
Smithsonian Institution Press, Washington, DC
Herrero-Pe?rezrul MD, Reyes-Bonilla HR, Garc??a-Dom??nguez F,
Cintra-Buenrostro CE (1999) Reproduction and growth of
Isostichopus fuscus (Echinodermata: Holothuroidea) in the
southern Gulf of California, Mexico. Mar Biol 135:521?532
Hopper DR, Hunter CL, Richmond RH (1998) Sexual reproduc-
tion of the tropical sea cucumber, Actinopyga mauritiana
(Echinodermata: Holothuroidea) in Guam. Bull Mar Sci 63:1?9
Humann PW (1992) Reef creature identi?cation, Florida, Carib-
bean and Bahamas. New World Publications, Jacksonville, Fla.
Humason GL (1972) Animal tissue techniques. Freeman, New
York
Krishnaswamy S, Krishnan S (1967) A report on the reproductive
cycle of the holothurian Holothuria scabra Jaeger. Curr Sci
36:155?156
Kubota T, Tomari M (1998) Reproduction in the apodid sea cu-
cumber Polycheira rufescens: semilunar spawning rhythm and
sex change. J Mar Biol Assoc UK 78:249?267
Mosher C (1982) Spawning behavior of the aspidochirote holo-
thurian Holothuria mexicana Ludwig. In: Lawrence JM (ed)
Proc Int Echinoderm Conf Tampa Bay. Balkema, Rotterdam,
pp 467?468
Ong-Che RG (1990) Reproductive cycle of Holothuria leucospilota
Brandt (Echinodermata: Holothuroidea) in Hong Kong and the
role of body tissues in reproduction. Asian Mar Biol 7:115?132
Pires-Nogueira RL, Freire CB, Rezende CR, Campos-Creasey LS
(2001) Reproductive cycle of the sea cucumber Isostichopus
badionotus (Echinodermata) from Ba??a da Ilha Grande, Rio de
Janeiro, Brazil (abstract). In: Abstracts of the 9th international
congress of invertebrate reproduction and development,
Rhodes University, Grahamstown, South Africa, 15?20 July
2001
Ramofa?a C, Battaglene SC, Bell JD, Byrne M (2000) Repro-
ductive biology of the commercial sea cucumber Holothuria
fuscogilva in the Solomon Islands. Mar Biol 136:1045?
1056
Reichenbach N (1999) Ecology and ?shery biology of Holothuria
fuscogilva (Echinodermata: Holothuroidea) in the Maldives.
Bull Mar Sci 64:103?113
Reichenbach N, Nishar Y, Saeed A (1996) Species and size related
trends in asexual propagation of commercially important spe-
cies of tropical sea cucumbers (Holothuroidea). J World Aquat
Soc 27:475?482
Richards AH, Bell LJ, Bell JD (1994) Inshore ?sheries resources of
Solomon Islands. Mar Pollut Bull 29:90?98
Rodr??guez E, Almanza R, Alvarado R (1993) Situacio?n biof??sica y
ambiental de la Provincia de Bocas del Toro. In: Heckadon-
Moreno S (ed) Agenda Ecolo?gica y Social para Bocas del Toro.
Impresora Continental, Panama City, pp 55?72
Rodr??guez-Milliet E, Pauls SM (1998) Sea cucumber ?sheries in
Venezuela. In: Mooi R, Telford M (eds) Proceedings of the 9th
international echinoderm conference. Balkema, Rotterdam,
pp 513?516
Sewell MA (1992) Reproduction of the temperate aspidochirote
Stichopus mollis (Echinodermata: Holothuroidea) in New Zea-
land. Ophelia 35:103?121
Sewell MA, Chia FS (1994) Reproduction of the intraovarian
brooding apodid Leptosynapta clarki (Echinodermata: Holo-
thuroidea) in British Columbia. Mar Biol 121:285?300
Sloan NA (1985) Echinoderm ?sheries of the world: a review. In:
Keegan BF, O?Connor BDS (eds) Echinodermata. Proc 5th Int
Echinoderm Conf. Balkema, Rotterdam, pp 109?124
Smiley S, McEuen F-S, Cha?ee C, Krishan S (1991) Echinoder-
mata: Holothuroidea. In: Giese AC, Pearse JS, Pearse VB (eds)
Reproduction of marine invertebrates, vol VI. Echinoderms
and lophophorates. Boxwood, Paci?c Grove, Calif., pp 663?
750
Sokal RR, Rohlf FJ (1995) Biometry, 3rd edn. Freeman, New
York
van Veghel MLJ (1993) Multiple species spawning on Curac?ao
reefs. Bull Mar Sci 52:1017?1021
279