Marine Biology (2003) 142: 281–288 DOI 10.1007/s00227-002-0947-x

C. Ramofafia Æ M. Byrne Æ C.S. Battaglene Reproduction of the commercial sea cucumber Holothuria scabra (Echinodermata: Holothuroidea) in the Solomon Islands

Received: 26 September 2001 / Accepted: 18 January 2002 / Published online: 12 November 2002 Springer-Verlag 2002

Abstract Over a 3-year period (1996–1998), reproduc- tubule recruitment model described for holothurian tion of the commercial sea cucumber Holothuria scabra oogenesis. Continuous reproduction in H. scabra and (Jaeger, 1833) was investigated in the Solomon Islands prolonged availability of mature gametes would facili- to determine the spawning pattern and whether ga- tate use of this species for aquaculture. metogenesis is continuous or seasonal. The gonad con- sisted of a single cohort of tubules that developed uniformly. Macroscopic examination of the gonads re- vealed that mature gametes were present throughout the Introduction year. Individuals with gonads at different stages of ma- turity were present in most samples. Partly spawned Holothuria scabra (commonly known as sandfish) is an gonads were prevalent in females, whereas mature go- aspidochirote holothurian widely distributed in coastal nads were prevalent in males. The time at which the regions throughout the Indo-Pacific region (Conand peak gonad index was recorded differed among years. 1998). H. scabra is often found on inner reef flats and Although gametogenesis was continuous, with a poten- near estuaries, half buried in the silty sand during the tial for prolonged gamete release, a period of enhanced day and emerging at night to feed (Mercier et al. 1999). spawning occurred during the dry season, from Sep- Two colour morphs, black and grey, are known, and tember to December. Maximum gonad indices were genetic data have shown that these are a single species reached prior to and during this period of enhanced (Uthicke and Benzie 1999). H. scabra is among ap- spawning. Histology revealed that gametogenesis reini- proximately 20 aspidochirote species that constitute the tiated in partly spawned gonads, resulting in the pres- beˆ che-de-mer fishery in the Indo-Pacific and is among ence of gametes at different stages of development in the the most valuable species in the trade (Conand and gonad. The uniform growth of gonad tubules indicated Byrne 1993; SPC 1994). that H. scabra does not conform to the progressive A biannual spawning pattern for H. scabra is re- ported in India, the Philippines, New Caledonia and Communicated by G.F. Humphrey, Sydney Indonesia (Table 1). These studies also reported that mature H. scabra can be found year-round, raising the C. Ramofafia (&) Æ M. Byrne possibility that reproduction may be continuous. In Department of Anatomy and Histology F13, subtropical regions of Australia, by contrast, annual University of Sydney, NSW 2006, Australia reproduction is reported for H. scabra (Harriot 1980; E-mail: chris.ramofafia@ffa.int Morgan 2000a). A recent review of the biology of Tel.: +677-25090 H. scabra concluded that, despite the investigations Fax: +677-23296 conducted over its geographic range, further research is C.S. Battaglene required to gain a better understanding of the repro- ICLARM–The World Fish Center, P.O. Box 438, Honiara, Solomon Islands ductive cycle and spawning cues (Hamel et al. 2002). In particular, the suggestion that tropical H. scabra popu- Present address: C.S. Battaglene lations have continuous gametogenesis has to be as- University of Tasmania, Tasmanian Aquaculture and Fisheries Institute, sessed by histological examination of gonad condition Taroona, TAS 7053, Australia throughout the year: most other studies have relied on Present address: C. Ramofafia the gonad index method (Hamel et al. 2002). ICLARM–The World Fish Center, In this study, we investigated the reproduction of P.O. Box 438, Honiara, Solomon Islands H. scabra in the Solomon Islands. Our aims were to 282

Table 1 Holothuria scabra. Reproduction of Holothuria scabra in the Indo-Pacific. – not documented Latitude Location Methods used Study duration GI max Reproductive pattern Spawning period Reference years

27N Red Sea Field spawning – – – Includes June Mortensen observation 1937 13N Philippines GI and 2NoContinuous mature Increased activity Ong Che and histology obvious specimens year in June and Gomez 1985 peaks round October (biannual peaks assumed) 09N India GI 1 July and Biannual July–August (main) Krishnaswamy October October–November and Krishnan 1967 05S Indonesia GI and 1 April and Biannual mature April and November Tuwo 1999 histology November specimens year round 09S Solomon GI and 3 August and Continuous mature Enhanced activity Present Islands histology October specimens year from September study (females) round to December August (males) 20S New GI 2 or more years’ August and Biannual August–September Conand Caledonia data, annual December December–January 1993 cycle presented 27S Australia GI and histology 1.5 December Annual November–January Harriott 1980 27S Australia GI and histology 1.5 November Annual mature November–December Morgan 2000a specimens seasonal determine the breeding pattern of H. scabra, to identify H. scabra in each of these stages was determined for all samples the timing of gamete availability, and to determine collected from May 1996 to December 1998. To determine whether GI differed among months and years, the whether gametogenesis is continuous or seasonal. GI data of both sexes were combined and compared by analysis of variance (ANOVA). Factors were year (Y) and month nested in year [M(Y)]; sample size n=17, the minimum monthly sample size Materials and methods was recorded. Where monthly sample size was greater than 17, specimens were picked at random to achieve n=17. A second ANOVA was also carried out to determine whether Samples of Holothuria scabra were collected from Kogu Halingi the GI data of females and males differed. For this ANOVA, fac- Bay on Kohinggo Island, Vona Vona Lagoon, Western Province of tors were sex (S), year (Y) and months nested in year [M(Y)]; the Solomon Islands (810’S, 15711’E). Samples were collected by sample size was n=6. For the ANOVA models to be balanced, the snorkelling or by SCUBA diving from May 1996 to December data obtained in 1996 were excluded from the analyses, since only 1998. Each specimen was placed in a plastic bag containing sea six monthly samples were collected. The 1997 and 1998 November water, sealed and packed in a fish box (60 l capacity), for trans- data were also excluded from the second ANOVA, since the male portation to the ICLARM Coastal Aquaculture Centre (CAC) on sample size in 1998 was lower than n=6. Homogeneity of variance Guadalcanal Island (0918’S, 15947’E). was tested using Cochran’s test and the data were transformed to In the laboratory, a longitudinal incision was made along the log (X+1) to achieve homogeneity of variance. Student–Newman– dorsal surface of each specimen and the coelomic fluid and gonad Keuls (SNK) multiple-range tests were used for a posterori com- were removed. Drained body weight (dwt) and gonad weight (gwt) parisons of monthly GI results that differed. were measured to 0.01 g and the gonads were fixed in 7% formalin in sea water. Gonad index (GI) was determined for each animal as (wet gwt/wet dwt)·100. Gonad indices were determined for each monthly sample and a mean GI was calculated for females and for males. Results The five-stage maturity scale, based on tubule morphology (Ramofafia et al. 2000), was used to describe gonad development. A total of 845 Holothuria scabra were collected, The maturity stages, based on macroscopic examination of the tubules, included: stage I, indeterminate; stage II, growing; stage 414 females (x dwt=644.94 g, SE=11.28) and 428 males III, mature; stage IV, partly spawned; and stage V, spent. Each (xmacr; dwt=655.49 g, SE=11.87). Two specimens gonad was assigned to one of these stages. Tubule size was also with indeterminate gonads (x dwt=384.95 g, SE= measured to determine whether tubule biopsies would provide a 105.05) were collected in July 1996 and April 1998 and meaningful assessment of reproductive condition. For each gonad, 15 tubules were removed, and their length and diameter measured one specimen that lacked a gonad (=396.50 g; dwt) was to the nearest millimetre. collected in December 1996. The sex ratio did not differ The histology of gametogenesis was also documented for each significantly from unity (v=0.85; P>0.05). specimen. For histology, gonad tubules were dehydrated through graded ethanols, rinsed in histoclear and embedded in paraffin. The sections (6 lm thick) were stained with haemotoxylin and eosin. Gonad morphology Based on the staining response, the gonads were assigned a gametogenic stage. Five gametogenic stages were defined: recovery, growing, mature, partly spawned, and spent, similar to the stages The gonad of H. scabra was a single tuft of numerous used in other studies of sea cucumbers (Sewell and Bergquist 1990; branched tubules on the right side of the dorsal mesen- Tuwo and Conand 1992; Ramofafia et al. 2000). The percentage of tery. All the tubules developed uniformly through the 283 gametic maturity stages. The size, colour and shape of tubules were reduced in size and appeared flaccid. The tubules and numbers of branches of each tubule de- unspawned tubules were similar in appearance to that pended on the stage of gonad development. Gonad described for mature tubules. Phagocytes were evident in maturity stages are compared in Table 2. All stages, with the lumen of both tubule types and were particularly the exception of indeterminate stage I, were present in abundant in the unspawned tubules, indicating that the samples throughout the year. gametes present were being resorbed. In some ovaries, Stage I, indeterminate: indeterminate gonads were oocytes in the unspawned tubules were clearly being small and composed of white unbranched tubules broken down. Owing to the presence of spawned and (<10 mm in length). Sex could not be determined, even unspawned tubules within the gonads, tubule length and by microscopic examination of gonad squash prepara- diameter were highly variable. Examination of ovary tions. squash preparations revealed the presence of immature Stage II, growing: growing gonads increased in size oocytes both in spawned and unspawned tubules, indi- with length, ranging from 20 to 70 mm. Examination of cating that re-initiation of gametogenesis had occurred. gonad squash preparations revealed the presence of de- Immature oocytes were particularly conspicuous in veloping gametes. In growing gonads, the tubules began spawned tubules. to bifurcate forming a series of branches. The length of Stage V, spent: spent gonads were greatly reduced in the tubule branches varied within each gonad. Ovaries size (Table 2). Tubules were wrinkled and shrunken in and testes were orange and glossy white, respectively. appearance. The weight of these gonads ranged from Stage III, mature: mature gonads had large fecund <1 g to 40 g. Relict oocytes and unspawned sperm tubules (30–200 mm long) which dominated the coelo- were usually seen in gonads 10–40 g in weight. The mic cavity (Table 2). During the mature stage the gonad basis of the smallest gonads (<1 g) typically tubules developed additional branches. Ovaries were a had a brown colour due to the accumulation of brown red-orange colour and oocytes were apparent through bodies, residual deposits characteristic of sea cucum- the thin transparent tubule wall. Testes were creamy bers (Jans et al. 1996). Examination of squash prepa- white and the tubules took on a beaded appearance. rations of these gonads revealed that most tubules were Within each gonad the total length and diameter of the empty. tubules were relatively uniform. Examination of ovary squash preparations showed the presence of oocytes at different stages of development. However, fully grown Histology oocytes were predominant. Squash preparations of testes revealed the presence of highly active sperm. Oogenesis Stage IV, partly spawned: partly spawned gonads possessed a combination of large fecund tubules and Ovaries in the recovery or indeterminate stage (I) had tubules that had released their gametes. Spawned previtellogenic oocytes (<20 lm diameter) developing

Table 2 Five maturity stages in the reproductive cycle of Holothuria scabra based on morphology of the gonad tubules Maturity stage, sex Gonad wt Tubule

g Length Diameter Branching Condition Colour

mm mm

I: indeterminate < 5 <10 <0.3 0 Gametes not evident White II: growing Female 5–30 30–70 0.7–1.4 1–2 Developing oocytes (20–80 lm) Orange Male 5–30 20–60 0.5–1.3 1–2 Sperm developing White III: mature Female 40–200 40–160 1.0–3.0 2–3 Oocytes visible through transparent tubule Reddish (120–170 lm) orange Male 30–200 30–170 0.5–2.5 2–3 Tubules packed with sperm, tubules have Creamy beaded appearance white IV: partly spawned Female 30–120 30–140 0.5–2.10. 1–3 Tubules reduced, empty regions of lumen Reddish visible, phagocytes present in spawned orange and unspawned tubules Male 20–90 20–150 3–1.9 1–3 Spawned tubules reduced, unspawned Creamy tubules packed with sperm white V: spent Female 0.1–40 6–130 0.1–1.8 1–3 Tubules reduced and shrunken, Pale relict oocytes present orange Male 0.1–40 10–140 0.1–1.3 1–3 Unspawned sperm present in shrunken White and wrinkled tubules 284 along the germinal epithelium (Fig. 1A). These oocytes ovaries resulted in the presence of overlapping genera- were strongly basophilic. Growing stage (stage II) ova- tions of oocytes (Fig. 1F, G). The tubules of spent ries were characterised by active vitellogenesis (Fig. 1B). ovaries (stage V) had a large lumen largely devoid of The oocytes were in the early (20–40 lm diameter) and contents. A few relict oocytes were occasionally present mid (30–80 lm diameter) vitellogenic stages and were and phagocytes were also seen in some specimens increasingly eosinophilic. The oocytes developed within (Fig. 1H). a follicle formed by somatic cells. Mature ovaries (stage III) were densely packed with eosinophilic late-vitello- genic oocytes (120–170 lm diameter) which remained Spermatogenesis within their follicle (Fig. 1C, D). The germinal vesicle of these oocytes was prominent and occupied a central or The pattern of testes development was also divided into eccentric position near the basal protuberance (Fig. 1D). five stages (Fig. 2). Recovery stage (stage I) testes were The protuberance attached the oocyte to the germinal characterised by a layer of basophilic spermatocytes epithelium (Fig. 1D). Partly spawned ovaries (stage IV) lining the germinal epithelium. Testes development then were characterised by the presence of unspawned oo- progressed to the growing stage (stage II) which was cytes and aggregations of phagocytes in the lumen characterised by dense layers of spermatocytes in the (Fig. 1E–G). Unspawned oocytes were loosely arranged germinal epithelium (Fig. 2A, B). Invaginations of with vacancies in the lumen left by recent spawning the germinal epithelium projected towards the centre of (Fig. 1E). Renewed gametogenesis in partly spawned the tubule, increasing the surface area of the germinal epithelium. As testis growth progressed, spermatozoa

Fig. 1A–H Holothuria scabra. Oogenesis. A Recovering ovary with Fig. 2A–I Holothuria scabra. Spermatogenesis. A, B Late recover- pre-vitellogenic oocytes (arrow). B Growing ovary with active ing/growing testis with infolds of germinal epithelium lined with vitellogenesis with early (evo) and mid (mvo) vitellogenic oocytes. columns of developing spermatocytes (arrowheads and arrows). C C, D Mature ovary. Fully grown oocytes within follicles (f) have a Late-growing testis with accumulation of spermatozoa (sz) in the large germinal vesicle (gv) and are attached to the germinal lumen. Infolds of the germinal epithelium lined with columns of epithelium by the protuberance (arrowhead). E–G Partly spawned developing spermatocytes persist (arrows). D Early mature testis ovary with loosely packed unspawned oocytes and renewed with increased predominance of spermatozoa (sz) in the lumen. gametogenesis with pre-vitellogenic (pv) and vitellogenic oocytes Spermatocytes (sc) still evident. E Mature testis packed with (vo). Relict oocytes (ro) from previous spawning period persist. H spermatozoa (sz). F, G Partly spawned testis. The spermatozoa (sz) Spent ovary with wrinkled and shrunken ovarian tubule containing are less dense (F) and renewed gametogenic production of relict oocytes (ro) and oocytes being degraded by phagocytic (p) spermatocytes (sc) occurring (G). H, I Spent testis can contain activity (Scale bars in A, B 25 lm; in C 133 lm; in D 65lm; in E–H unspawned spermatozoa (sz)(H) or an empty lumen (I)(Scale bars 240 lm) in A, B 90 lm; C, D, G 110 lm; E 750 lm; F, H, I 150 lm) 285 accumulated in the lumen (Fig. 2C, D). In mature testes females were recorded in October 1997 and August 1998 (stage III), the lumen was packed with spermatozoa and (Fig. 3A). The particularly high mean GI value obtained the infolds of the germinal epithelium were reduced or in October 1997 was attributable to four specimens (total absent (Fig. 2E). The wall of the mature testes was also n=10) with unusually large gonads, with wet weights at its minimal thickness. In partly spawned (stage IV) ranging from 125 to 210 g. The weight of gravid ovaries testes the spermatozoa were less densely packed encountered during the 3 years of sampling more typi- (Fig. 2F, G) and phagocytes were also present in the cally ranged between 70 and 100 g. In 1997 and 1998, the lumen. In some testes, gaps in the spermatozoal mass maximum GI values in the females were followed by a indicated the occurrence of gamete release. Gameto- decrease to December in 1997 and to October in 1998. genesis reinitiated in partly spawned testes, as indicated The males recorded the largest GIs from August to by the presence of developing spermatocytes and the October 1997, but exhibited no similar pattern in 1998. reappearance of infolds of the germinal layer (Fig. 2G). The lowest GIs for both sexes were recorded in July 1997. Spent (stage V) testes were largely devoid of contents, The results of the ANOVA comparing GI data re- with the exception of a few unspawned spermatozoa vealed that the GI differed significantly among months (Fig. 2H, I). In some spent testes, the germinal epithe- and that this depended upon year (Table 3, A). This was lium took on a convoluted appearance but renewed indicated by the nested term [M(Y)] being significantly gametogenesis was not observed. different (Table 3, A). The SNK tests showed that the October GI differed from that of all other months in 1997 and that the August GI differed from that of all Gonad index and tubule size other months in 1998. In the second ANOVA, com- paring the GI for females and males, the factor S·M(Y) The pattern of gonad growth, as exhibited by the GI was also significantly different (Table 3, B) indicating data, was variable between sexes and across years that the GI and the reproductive activity of each sex (Fig. 3A). For most of the study, the GI remained be- were also variable among months and between years. tween 2 and 6, with the GIs of female H. scabra often The pattern of the mean tubule size was variable in being higher than those of males. The largest GIs for both sexes and not synchronised with the growth in

Fig. 3A–C Holothuria scabra. A Gonad index for females (closed circle) and males (open circle) from Kogu Halingi Bay, Solomon Islands over a 3-year period. B tubule length and C diameter of females (closed cir- cle) and males (open circle) over the same period (Bars are stan- dard error of the mean) 286

Table 3 ANOVA of the GI data; ns not significant; ** P £ 0.01 n=123) were the most frequently encountered maturity A. Combined female and male GI data stages in females. Females with growing, mature or Sources of variation df MS FPrecovering ovaries were comparatively rare (Fig. 4A). Year Y 1 0.29 0.24 ns In male H. scabra, individuals with mature (33%, Month (Y) 20 1.20 4.42 ** n=143), growing (29%, n=124), and partly spawned Residual 352 0.27 testes (25%, n=108) were common (Fig. 4B). Like the Total 373 females, the presence of testes at three or more stages of B. Individual female and male GI data development also indicated that gametogenesis was Sources of variation df MS FP Sex S 1 0.33 0.53 ns asynchronous among the males. Spent testes (12%, Year Y 1 0.32 0.45 ns n=53) were less frequent than spent ovaries. The re- Month (Y) 20 0.71 3.14 ** covery stage was only seen in one male collected in S·Y 1 0.61 0.26 ns December 1996 (Fig. 4B). S·Month (Y) 20 0.62 2.73 ** Residual 220 0.23 The data from gonad index, tubule size and histology Total 263 suggested that H. scabra has a continuous and asyn- chronous pattern of gonad development in the Solomon Islands (Fig. 3, 4). However, reproductive activity did weight of the gonad as measured by GI data (Fig. 3). exhibit some consistency among months, with maximum For instance, in 1997, ovary tubule length was greatest gonad growth for the females recorded in the August to in August, two months before the GI maximum was October period in 1997 and 1998. This was followed by a recorded. Conversely, the ovary tubule diameter was decrease in GI due to gamete release. In September and greatest in August 1997, corresponding to the peak GI. October 1996, some of the H. scabra underwent spon- In 1998, the longest tubules were recorded in March, taneous gamete release before the gonads could be used several months before the maximum GI occurred. The for analysis. pattern of testis tubule length and diameter-change paralleled that for the ovary (Fig. 3B,C). Discussion

Reproductive cycle Holothuria scabra is one of the most widely distributed tropical sea cucumbers, with a distribution encompass- Gametogenesis was asynchronous in H. scabra and did ing 54 degrees of latitude (Table 1). Over this broad not exhibit a seasonal pattern (Fig. 4). For example, distribution, H. scabra exhibits two main basic repro- most samples contained females with ovaries in at least ductive patterns: seasonally predictable spawning at three different stages of oogenesis (Fig. 4). Overall, high latitudes; and aseasonal spawning at low latitudes partly spawned (53%, n=220) and spent ovaries (30%, (Table 1). Reproduction of H. scabra in the Solomon

Fig. 4A,B Holothuria scabra. Gonad histological stage in gametogenic cycle from A fe- males and B males collected from Kogu Halingi Bay, May 1996 to December 1998. In September and October 1996 all specimens spawned, so their gonads were not used for his- tology. (n size of sample exam- ined each month) 287 Islands was characterised by continuous, asynchronous mental factors, such as the lunar cycle which appears to gametogenesis, with mature gametes present year round. stimulate aggregation and spawning in H. scabra around Spawning appeared to be episodic through the year, the new moon (Hamel et al. 2002). Synchronous with a period of enhanced activity from September to spawning of H. scabra and other aspidochirotes in the December. In the Philippines, H. scabra exhibits two field and in captivity has also been noted to coincide periods of enhanced spawning superimposed on con- with different lunar phases, depending on location tinual reproductive maturity (Ong Che and Gomez (Shelley 1982; Babcock et al. 1992; Morgan 2000b; 1995). The continuous gametogenesis that characterised Battaglene et al. 2002). Hamel et al. (2002) provide H. scabra in the Solomon Islands and the Philippines further details of the influence of various environmental may extend broadly in the Indo-Pacific region (20Nto factors and the timing of spawning. 20S; Table 1). At the northern and the southern limits Asynchronous gametogenesis and the year-round of its range, H. scabra has an annual pattern of syn- availability of mature gametes in H. scabra is not typical chronous gametogenisis (Mortensen 1937; Morgan of other tropical aspidochirotes. For example, H. fus- 2000a) potentially cued by seasonally predictable fac- cogilva and Actinopyga mauritiana have synchronous and tors, such as day length and water temperature, at higher seasonal gonad development, suggested to be cued by latitudes. The reproductive strategies in H. scabra, seasonally predictable changes in environmental factors therefore, support the hypothesis that reproduction of (Ramofafia et al. 2000, 2001a). Further work is required widely distributed tropical marine invertebrates changes to determine the factors that modulate gametogenesis in from continuous to seasonal with increasing latitude H. scabra in tropical regions. Gametogenic renewal oc- (Thorson 1950). curred in partly spawned gonads in H. scabra, resulting Across the Indo-Pacific region, spawning in H. scabra in the presence of gametes of different stages of devel- is temporally variable, with biannual and continuous opment in the gonads. This renewed gametogenesis in H. activities recorded (Table 1). Even when twice-yearly scabra provided the mechanism for continuous spawn- spawning peaks have been observed, the months differed ing. In contrast, the fate of new gametes in partly among years (Table 1). Furthermore, the lack of an spawned tubules is not known for H. fuscogilva, A. annually predictable GI pattern, as seen in the Solomon mauritiana and several other holothurians, because the Islands (the present study) and in the Philippines (Ong gonad tubules subsequently become reduced or are re- Che and Gomez 1985), also suggests that gonad growth sorbed completely (Ramofafia et al. 2000, 2001a). Total can be temporally variable. This interannual variability gonad resorption is comparatively rare in H. scabra in in reproduction of populations of H. scabra in the Indo- the Solomon Islands and difficult to distinguish macro- Pacific region (Table 1) may be influenced by variable scopically from animals that may have eviscerated. local environmental conditions, particularly those asso- Partly spawned gonads were common in female H. ciated with the coastal mangrove and sea-grass habitats scabra, whereas mature gonads were common in males. occupied by this species. Spawning of H. scabra in the Owing to the presence of gonads at different stages of region may be opportunistic, with some flexibility to maturity in the population throughout the year, exam- respond to ephemeral favourable conditions. ination of gonad tubule biopsies from a few H. scabra The presence of black and grey morphs of H. scabra can not be used as a tool to assess the reproductive cycle, across the Indo-Pacific could be taken to suggest that as has been done for other species (Ramofafia et al. variable reproduction in this species has an underlying 2000, 2001a). None the less, the biopsy method may be genetic influence. However, there were no genetic dif- useful for selective harvesting of mature individuals in ferences between these morphs in Australia (Uthicke the field for use in breeding programmes. and Benzie 1999). None the less, genetic differences be- The gonad in H. scabra is a single tuft of tubules at a tween populations in geographical regions were large, similar stage of development, a feature characteristic of compared to other marine broadcast spawners (Uthicke tropical holothurians (Conand 1981, 1993; Ramofafia and Benzie 2001). et al. 2000; 2001a). Such uniform development of tu- Although H. scabra spawns year round, the presence bules differs from the findings of Smiley (1988) and of a period of enhanced spawning (September to Tuwo and Conand (1992) for the temperate aspidochi- December) was consistent over the 3 years of study in rotes H. forskali and Stichopus californicus, respectively. the Solomon Islands, suggesting some role for external In these two species, the gonad comprises distinct co- cues. This period of spawning coincided with the dry horts of tubules that are progressively recruited. Smiley season, before the monsoon begins in late December. It (1988) suggested that this type of ovary development also coincided with periods of increased day length and (tubule recruitment model) was characteristic for Holo- water temperature (Ramofafia et al. 2000; 2001a). thuroidea. However, results for H. scabra and other Temperature, salinity and photoperiod have been sug- holothurians demonstrate that oogenesis in most tropi- gested to influence the biannual spawning pattern of H. cal and temperate holothurians does not conform to this scabra in India, the Philippines, New Caledonia and model (Sewell et al. 1997; Ramofafia et al. 2000; 2001a; Indonesia (Krisnaswamy and Krishnan 1967; Ong Che Ramofafia and Byrne 2001). and Gomez 1985; Conand 1993; Tuwo 1999). Gamete The ability to breed year round is a great advantage release is also likely to be influenced by other environ- for the aquaculture of H. scabra. Moreover, spawning 288 induction is relatively easy in mature H. scabra com- intracoelomic brown bodies in the sea cucumber Holothuria pared to other tropical aspidochirotes (Ramofafia et al. tubulosa. Cell Tissue Res 283:99–106 Krishnaswamy S, Krishnan S (1967) A report on the reproductive 1995; Morgan 2000b; Battaglene et al. 2002). In agree- cycle of the holothurian Holothuria scabra Jaeger. Curr Sci ment with our findings, spawning can be induced in H. 36:155–156 scabra year-round in the Solomon Islands with the Mercier A, Battaglene SC, Hamel J-F (1999) Daily burrowing cycle greatest response observed from September to December and feeding activity of juvenile sea cucumbers Holothuria scabra in response to environmental factors. J Exp Mar Biol Ecol (Battaglene et al. 2002). H. scabra is relatively easy to 239:125–156 culture compared with other tropical holothurian spe- Morgan AD (2000a) Aspects of the reproductive cycle of the sea cies, and outplanted juveniles survived well (Battaglene cucumber Holothuria scabra (Echinodermata: Holothuroidea). et al. 1999; Ramofafia et al. 2001b). Our assessment of Bull Mar Sci 66:47–57 gonad development in H. scabra, together with results of Morgan AD (2000b) Induction of spawning in the sea cucumber Holothuria scabra (Echinodermata: Holothuriodea). J World other studies (Table 1), leads us to suggest that contin- Aquacult Soc 31:186–194 uous and asynchronous reproduction as seen here for Mortensen TH (1937) Contributions to the study of the develop- the species in the Solomon Islands may be characteristic ment and larval forms of echinoderms. III. R Acad Sci Lett Den of this species in the tropical Indo-Pacific. (Ser 9) 4(1):1–39 Ong Che RG, Gomez ED (1985) Reproductive periodicity of Holothuria scabra Jaeger at Calatagan, Batangas, Philippines. Acknowledgments We thank the staff of the ICLARM Coastal Asian Mar Biol 2:21–30 Aquaculture Centre and Nusa Tupe field station for assistance; Ramofafia C, Byrne M (2001) Assessment of the ‘‘tubule recruit- particularly, Maxwell Saurongo, Joseph Olisia, Evizel Seymour ment model’’ in three tropical aspidochirote holothurians. SPC and Idris Lane. We are grateful to Thomas Tebounaba and Ste- Inf Bull Beˆ che-de-mer 15:13–16 phen Sibiti for broodstock collection. The Electron Microscope Ramofafia C, Gervis M, Bell JD (1995) Spawning and early larval Unit, University of Sydney, assisted with photography. Funds were rearing of Holothuria atra. SPC Inf Bull Beˆ che-de-mer 7:2–5 provided by the Australian Centre for International Agricultural Ramofafia C, Battaglene CS, Bell JD, Byrne M (2000) Reproduc- Research, an AusAID scholarship (to CR) and an ARC grant (to tive biology of the commercial sea cucumber Holothuria fus- MB). Jean-Francois Hamel, Annie Mercier and Cletus Pita pro- cogilva in the Solomon Islands. Mar Biol 136:1045–1056 vided fruitful discussions throughout the study. The manuscript Ramofafia C, Battaglene CS, Byrne M (2001a) Reproduction of the was improved from comments by Johann Bell and two reviewers. intertidal sea cucumber Actinopyga mauritiana in the Solomon This study is ICLARM contribution no. 1639. Islands. J Mar Biol Assoc UK 81:523–531 Ramofafia C, Battaglene CS, Byrne M (2001b) Larval development in the tropical sea cucumber Holothuria scabra. In: Barker MF References (ed) Echinoderms 2000. Swets and Zeitlinger, Lisse, pp 369–375 Sewell MA, Bergquist PR (1990) Variability in the reproductive cycle of Stichopus mollis (Echinodermata: Holothuroidea). Babcock R, Mundy C, Keesing J, Oliver J (1992) Predictable and Invertebr Reprod Dev 17:1–7 unpredictable spawning events: in situ behavioural data from Sewell MA, Tyler PA, Young CM, Conand C (1997) Ovarian free-spawning coral reef invertebrates. Invertebr Reprod Dev development in the Class Holothuroidea: a reassessment of the 22:213–228 ‘‘tubule recruitment model’’. Biol Bull 192:17–26 Battaglene SC, Seymour EJ, Ramofafia C (1999) Survival and Shelley C (1982) Aspects of the distribution, reproduction growth growth of cultured juvenile sea cucumbers, Holothuria scabra. and fishery potential of holothurians (beˆ che-de-mer) in the Aquaculture 178:293–322 Papuan coastal lagoon. MSc thesis. University of Papua New Battaglene SC, Seymour EJ, Ramofafia C, Lane I (2002) Guinea, Port Moresby, Papua New Guinea Spawning induction of three tropical sea cucumbers, Holoth- Smiley S (1988) The dynamics of oogenesis and the annual ovarian uria scabra, H. fuscogilva and Actinopyga mauritiana. Aqua- cycle of Stichopus californicus (Echinodermata: Holothuroidea). culture 207:29–47 Biol Bull 175:79–93 Conand C (1981) Sexual reproduction of three commercially im- SPC (1994) Sea cucumbers and beˆ che-de-mer of the tropical portant holothurian species (Echinodermata) from the lagoon Pacific. A handbook for fishers. South Pacific Commission, of New Caledonia. Bull Mar Sci 31:523–543 Noumea, New Caledonia Conand C (1993) Reproductive biology of the holothurians from Thorson G (1950) Reproduction and larval ecology of marine the major communities of the New Caledonian Lagoon. Mar bottom invertebrates. Biol Rev 25:1–45 Biol 116:439–450 Tuwo A (1999) Reproductive cycle of the holothurian Holothuria Conand C (1998) Holothurians. In: Carpenter K, Niem V (eds) scabra in Saugi Island, Spermonde archipelago, southwest FAO species identification guide. The marine living resources of Sulawesi, Indonesia. SPC Info Bull Beˆ che-de-mer 11:9–12 the western central Pacific, vol 2. Cephalopods, crustaceans, Tuwo A, Conand C (1992) Reproductive biology of the holothu- holothurians and sharks. FAO, Rome, pp 1158–1189 rian Holothuria forskali (Echinodermata). J Mar Biol Assoc UK Conand C, Byrne M (1993) A review of recent developments in the 72:745–758 world sea cucumber fisheries. Mar Fish Rev 55:1–13 Uthicke S, Benzie J (1999) Allozyme variation as a tool for beˆ che- Hamel J-F, Pawson DL, Conand C, Mercier A (2002) The sea de-mer fisheries management: a study on Holothuria scabra cucumber Holothuria scabra (Holothuroidea: Echinodermata): (sandfish) SPC Info Bull Beˆ che-de-mer 12:18–23 its biology and its exploitation as beˆ che-de-mer. Adv Mar Biol Uthicke S, Benzie J (2001) Restricted gene flow between Holothuria 41:131–233 scabra (Echinoderm: Holothuroidea) populations along the Harriott VJ (1980) The ecology of holothurian fauna of Heron north-east coast of Australia and the Solomon Islands. Mar Reef and Morton Bay. MSc thesis. University of Queensland, Ecol Prog Ser 216:109–117 Australia Jans D, Dubois P, Jangoux M (1996) Defensive mechanisms of holothuroids (Echinodermata): formation, role, and fate of