Larval and spat culture of the Western Australian silver- or goldlip pearl oyster, Pinctada maxima Jameson (Mollusca: Pteriidae)

Aquaculture ELSEVIER

Aquaculture 126 (1994) 35-50

Larval and spat culture of the Western Australian silver- or goldlip pearl oyster, Pinctadu maxima Jameson (Mollusca: Pteriidae) Robert A. Rose*, Shane B. Baker’ Fishen’es Department,

Western Australian Marine Research, Laboratories, WA 6020, Australia

P.O. Box 20, North Beach,

Accepted 1 April 1994

Abstract The larvae of Pinctada maxima were fed Tahitian Isochrysis galbana, Chaetoceros calcitrans, C. gracilis and Nannochloropsis oculata. Food concentrations for larvae were increased gradually from 0.2-2 algal cells. ~1~’ on day 1 to 20-40 cells * ~1~i on day 30. Initial stocking densities of l-8 larvae.ml-’ were reduced to 0.5-l larvae-ml-’ at settlement. For 70% of the population, settlement began on day 24 and lasted up to 7 days. Ten to 15% of the population failed to grow appreciably and another lO-15% grew comparatively rapidly, reaching settlement by days 15-17. The smallest spat observed on day 28 was 33 1 pm shell length (SL) X 305 pm shell height (SH). Spat were fed the same phytoplankton as the larvae, as well as Tetraselrnis chuii twice daily at 40-285 cells ,x1- ’ over 5 months. Spat reared in downwellers at densities of 4 and 25 individuals. 100 cm-’ grew 9.6 and 6 mm.mthh’ SH, respectively. Those reared in plastic cages at sea at densities of 3 and 7 individuals. 100 crnm2 grew 9.2 and 7.3 mm mth-’ SH, respectively. Mortality 5 months after settlement was l-2% for those reared in the hatchery and 9-12% for those reared at sea. Hatchery-propagated spat were similar in appearance to natural spat and 20-25% were suitable for pearl culture 19 months after fertilization, or when they were 120 mm SH. Keywords:

Pearl oyster; Spawning; Larvae; Spat; Growth; Survival

1. Introduction The Western Australian Pearling Industry, estimated to be worth $A 125 ($US 87) million per annum in 1991, relies entirely on wild stocks of the silver- or gold-lip pearl * Corresponding author, present address: Pearl Oyster Propagators Pty. Ltd., 4 Daniels Street, Ludmilla, NT 0820, Australia. ’ Present address: 15 Ranger Trail, Edgewater, WA 6027, Australia. 0044~8486/94/$07.00 0 1994 Elsevier Science B.V. All rights reserved SsO10044-8486(94)00114-4

36

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

oyster, Pinctada maxima (Jameson), for pearl production. The oysters are cultured for round, silver-white pearls 10 mm or greater in diameter ( “south sea pearls”). Older, larger animals no longer suitable for round pearl culture are seeded for half pearls. Approximately 1 year later they are harvested for the pearls, mother-of-pearl shell and adductor-muscle meat. Since 1982 the fishery in Western Australia has been subjected to a quota management system to control exploitation and pearl production (Dybdahl and Rose, 1986). Exclusive reliance on wild stocks places this industry at economic risk since bivalve populations frequently exhibit wide fluctuations in levels of recruitment. The extent of this risk would increase should industry experience unexpectedly high levels of mortality amongst oysters after collection as occurred from mid-1970s to the mid- 1980s (Pass et al., 1987). In addition, the costs of collection and transportation of wild oysters for culture purposes are continually increasing and are at present between A$19 and 20 ($US 12.8-13.5) per oyster in Western Australia and between A$ 17 and 27 ($US 11.5-18.2) in Queensland and the Northern Territory. Any sustainable, cost-effective expansion of production in the future is unlikely to occur unless hatchery-propagated oysters become available as an alternative source to wild stock. Furthermore, via hatchery propagation, large cohorts of genetically selected individuals could be used for pearl cultivation. This would improve the uniformity and quality of the pearls and thus the price since pearIs are generally soId in allotments of similar weight, shape, colour, lustre, and surface complexion. Research on the biology, larval and pearl culture of P. maxima in Australia began during the middle-late 1950s at C.S.I.R.O.‘s Thursday Island Field Station, Queensland. However, none of this research relating to larval culture was successful or published. The only nonJapanese research published during the 1960s and 1970s was that of Minaur ( 1969) from the Cape York Pearling Company, Thursday Island. Unfortunately, this project failed to successfully rear the post-larvae beyond settlement and was discontinued. Concurrently, Japanese scientists were also developing the technology to propagate P. maxima. During the early 1960s and by the late 1970s to mid-1980s they had succeeded in settling P. maxima on a large scale at several private hatcheries located in Australia and southeast Asia (e.g., Torres Strait, Queensland: Kuri Bay, Western Australia; Amami O’Shima, Japan; Sabah, Malaysia; and Sumatra, Indonesia). By 1991,4 or 5 commercial hatcheries were operating in Australia and staffed by either Japanese or Australian technicians (two in Western Australia, one in the Northern Territory, and one or two in Queensland). However, except for a Japanese publication by Tanaka and Kumeta ( 198 1) on their work in the Torres Strait, these achievements have remained largely confidential. This paper describes the successful breeding of P. maxima at the W.A. Fisheries Research Station, Broome, Western Australia during the 1987-1988 and 1988-1989 seasons. The descriptions of the larvae and spat are derived from 3 spawnings and are the first for P. maxima from Western Australia. Pinctada maxima, which grows to 270 mm or more in shell height (SH) and weighs up to 5.5 kg, is the largest of the three Pinctada species cultured for round pearls, the other two being P. margaritifera (Linnaeus) and P.fucata (Gould). Predominantly a sublittoral bivalve found at depths as great as 50 m, P maxima inhabits subtropical and tropical coastal waters throughout southeast Asia and northern Australia (Wada, 1953a; Hynd, 1955).

R.A. Rose, S.B. Baker/Aquaculture 126 (1994) 35-50

31

A protandrous hermaphrodite (Wada, 1953a; Rose et al., 1990)) individuals of P.maxima generally mature as males during the first year of life ( 120 mm SH).As they grow (age), the incidence of female sexuality increases so that the sex-ratio in wild populations approaches 1 to 1 amongst individuals which are 170 mm SH or greater, or weigh 1 kg or more (Wada, 1953a; Rose et al., 1986). In northern Australia, P.maxima breeds from September/October (spring) to April/May (autumn), with the main period occurring from mid-October to December and a secondary period occurring in February to March (Wada, 1953a; Tanaka and Kumeta, 1981; Rose et al., 1990).

2. Materials and methods Spawnings

Progeny from one field and two hatchery spawnings were reared at a temporary, research hatchery located in a storage shed at the end of the Broome Jetty (lat. 18”OO’ S; long. 122’13’ E) . Adults of P.maxima used for hatchery spawnings measured 150 mm SH or greater and were collected at depths of 25-30 m from fishing grounds off Eighty-Mile Beach and Gantheaume Point, Broome, Western Australia. Oysters collected during 1987 were held in 6-pocket nets positioned near the surface under the Broome Jetty for 5 weeks before inducing them to spawn on 22 October. Those collected in 1988 were brought back to the hatchery and spawned within 6 days after collection on 28 November. In the field, zygotes were obtained fortuitously from spawnings produced by seeded, commercial oysters held in carrying tanks on board vessels in transit from the fishing grounds to pearl culture farms on 15 November, 1988. Broodstock used for hatchery spawnings were cleaned and placed into a tank containing gently aerated, 1 pm filtered, ultraviolet-irradiated 35 ppt seawater and subjected to thermal stimulation by heating the seawater from 26 to 3 1°C (Rose et al., 1986). During the October, 1987 spawning, 5 females and 7 males out of a total of 57 oysters produced 11.29 X lo6 zygotes 4 h after the beginning of the spawning trial. During November 1988, 4 females and 8 males out of a total of 18 oysters spawned spontaneously and vigorously 15-30 min after being placed in the spawning tank and exposed to 10 pm filtered seawater heated l2°C above ambient seawater temperature of 28°C. After 2 h these oysters had produced 47 X lo6 zygotes. In the field, a total of 14.8 X lo6 zygotes were collected from 4 500 oysters and these zygotes were brought back to the hatchery for subsequent culturing. Larval and spat culture

Procedures for rearing the larvae were similar to those described for P.maxima by Minaur ( 1969) and Tanaka and Kumeta ( 1981)) P.fucata by Alagarswami et al. ( 1983) and P. margaritiferu by Alagarswami et al. (1989). After fertilization, zygotes were siphoned from the spawning tank, poured through 100 and 80 pm nylon sieves to remove gonadal debris, faeces and pseudofaeces, collected onto a 20 or 25 pm sieve and washed with submicron-filtered seawater. Zygotes were then counted and transferred to 30-litre hatching bins filled with submicron or 1 pm filtered seawater. Approximately 18 h after fertilization, only newly formed straight-hinged veligers or D-shaped larvae swimming in the water

38

R.A. Rose, S.B. Baker/Aquaculture

I26 (1994) 35-50

column were siphoned into 20-litre bins, counted and poured into 30-, loo-, 500- or 1 OOOlitre tanks filled with 1 or 5 pm filtered seawater. The seawater temperature ranged from 25 to 295°C and salinity from 35 to 36 ppt. The number of complete water changes prior to settlement depended on the initial stocking density, with lower densities of 1 larva*ml-’ requiring only 1 change on day 10 and higher densities of up to 5-8 larvae*ml-’ requiring a change every 5 or 6 days (a total of 4-5). Densities at settlement and metamorphosis were kept at 0.5-l larva*ml- ’ or less. Larvae and post-larvae were fed a mixture of unicellular phytoplankton (Tahitian Zsochrysis galbana, Nannochloropsis oculata, Chaetoceros calcitrans and C. gracilis) with bacterial levels less than 3 X 10’ bacteria- ml- ‘. On several occasions the antibiotics chloramphenicol (0.5 mg +ml - ’ ) and streptomycin sulphate (0.5 mg ml - ’ ) ) were added to the algal cultures to keep the bacterial load below this level. Food concentrations were monitored daily and, depending on the larval stocking density, were increased gradually from 0.2-2 algal cells. z.K’ on day 1 to 20-40 cells Z-K’ on day 30-31. Larvae reared in October, 1987 were fed once daily. Those reared in November, 1988, were fed once daily up to day 4 and twice daily thereafter, with half of their total ration in the morning and the other half in the evening. Culture tanks were gently aerated during settlement and post-larvae or plantigrades were provided with collectors made out of monofilament nylon fishing line, dark plastic shade cloth, dark glass or grey plastic plates. Spat which had settled onto the sides and bottoms of their culture tanks were transferred onto dark glass plates or shade cloth 2 weeks after settlement. Spat which had attached to glass plates were moved to downweller units 3 weeks after settlement and stocked at densities ranging from 4 to 25 spate 100 cm-‘. Plantigrades and newly metamorphosed spat were fed Tetraselmis chuii as well as the above species of phytoplankton. Spat were fed twice daily on a mixed algal diet which increased from 40 to 285 algal cells* ~1~ ’ over 5 months. Several thousand spat attached to monofilament fishing line were transferred to grow-out culture 3 weeks after settlement and placed in galvanized wire trays lined with removable plastic mesh. One or 2 months after settlement, spat in the downwellers were routinely removed and placed in circular, plastic-mesh trays and surface-cultured at sea. These spat were stocked at densities ranging from 3 to 7 individuals* 100 cm-‘. To record development, 30-60 larvae were measured from each spawning, fixed and preserved periodically with a working solution of triple fix (Chanley, 1981). Sampled larvae were measured and photographed using a compound microscope with attached camera. The provincular structure of prodissoconchs of umbonal larvae were examined with a Philips 501-B scanning electron microscope. The valves of the larvae were cleaned with 12% (v/v) sodium hypochlorite solution, washed in distilled water, mounted with double-sided adhesive tape onto stubs and coated with 90 nm of gold (Rose et al., 1988). Descriptive terminology of larval stages and provincular structures follows that adopted by Loosanoff et al. ( 1966) and Rees ( 1950), respectively. Paired shell dimensions measured were length (antero-posterior distance parallel to hinge: SL) and height (dorso-ventral distance from hinge-line or umbo to ventral margin of shell: SH) . Unless specified, growth rate of the spat shell is given as SH since this dimension is used to determine the legal minimum size of commercial pearl oysters. Growth curves for larvae and spat were derived l

R.A. Rose, S.B. Baker/Aquaculture

I26 (1994) 35-50

39

from over 300 measurements pooled from all 3 spawnings. In order to clearly display these curves, not all of the data have been plotted for each figure presented.

3. Results

Development The colour of female gonads of P. maxima is orange or yellow and that of males is white. Females induced to spawn each released between 2 X lo6 and 12 X lo6 golden-yellow, spherical ova with little vitelline material. The diameter of 50 ova from all 3 spawnings was 60 ,um (Fig. 1A). The pattern of cleavage and embryogenesis in P. maxima was typical of bivalves with planktotrophic development and has been described and illustrated in detail by Wada ( 1953b). At temperatures ranging from 27 to 29°C first cleavage began with the formation of the first polar lobe 15 min after expulsion of the 2nd polar body and was completed 40 min after fertilization. Second cleavage (4-cell stage) occurred within 1 h (Fig. 1A) and morula within 3 h. A ciliated, rotating gastrula (Fig. 1B) and trochophore (Fig. 1C) displaying an apical flagella and shell gland were present 5 and 7 h after fertilization, respectively. The straight-hinged veliger or D-shaped larva appeared after 18-24 h (Fig. 1D) but generally did not show a fully developed viscerum until after 26 h. The prodissoconch I (or embryonic) shell of day 1 veligers averaged 79 pm SL and 67 pm SH. Growth and development of the veligers was uniform up to day 6, but beyond this day differences were observed. Seventy to 80% of the straight-hinged veligers had developed a rudimentary umbo on each valve by day 8-9 or when their shell averaged 110 w SL X 100 SH pm (Fig. 1E). By day 10, or when larvae were 114 pm SL X 103 pm SH, the umbones were fully formed (Fig. 1F). At this stage, concentric growth lines could be readily observed at the shell margin and the hinge structure of the shell had a pink or purplish tinge. Of the remaining 20-30% of the straight-hinged veligers, half grew relatively slowly, failing to reach the umbonal veliger stage until day 16. In contrast, the other half ( lO-15%) began developing umbones on day 6 when they ranged from 103 pm SL X 88 pm SH to 135 pm SL X 125 pm SH. These larvae continued to develop and grow comparatively more rapidly than the others so that by days 12-14, 10% of the fastest growers (averaging 253 pm SL X 218 pm SH) were between 2.5 and 3 times larger than the slowest growers (95 pm SL X 90 ,um SH) (Fig. 1G). Rapidly developing veligers reached the pediveliger stage (displaying an eye spot and a fully formed foot) on days 12- 14 and settled on days 1517, when they ranged from 270 pm SL X 220 pm SH to 307 pm SL X 269 pm SH. The percentage of crawling pediveligers during this period was typically less than 5% of the total population sampled. On day 19, 85-90% of the larvae were umbonal veligers, averaging 156 pm SLX 144 pm SH. By day 20, a faint red-pigmented eye spot (5-7 pm in diameter) could be observed embedded in the velum on the side near the rudimentary foot in veligers 205 pm SL X 175 pm SH or greater (Fig. 1H). On days 22-24, the pediveliger was the dominant stage (70%)) averaging 211 pm SL X 188 pm SH on day 22 and 230 pm SL X 200 ,um SH on day 24 (Fig. 1 I). During this period, those which were 230 pm SL or greater possessed

40

R.A. Rose, S. B. Baker /Aquaculture 126 (1994) 35-50

R.A. Rose, S.B. Baker/Aquaculture I26 (1994) 35-M

41

Fig. 2. Scanning electron micrographs of the surface and hinge structure of the larval shells of Pin&z& maxima. (A) Surface of the right valve of a 23-day-old veliger showing distinct growth rings (gr) . (B) Both valves of a 23-day-old veliger showing embryonic or prodissoconch I shell (PI) and larval or prodissoconch II shell (pII). (C) Inside view of the hinge on a pair of valves from a 23-day-old umbonal veliger. (D) Provincular structure of the left valve of a 24-day-old veliger showing taxodontal teeth (t), provinculum ( pr) and flange ( fl) .

two dark-red eye spots ( 10 pm in diameter), one spot near the proximal end of each side of a fully developed foot. The majority of pediveligers began metamorphosing into plantigrades (post-larvae) on

Fig. 1. Embryonic, larval and early spat development of Pinctada maxima. (A) Early cell division 1 h after fertilization: ovum (0) ; early first cleavage ( 1C) ; second cleavage (2C) with embryo displaying 3 out of 4 cells and a polar lobe ( pl) . (B) Gashula h after fertilization. (C) Eight-hour-old trochophore showing shell formation (sh) and apical flagella (af) . (D) Day-old straight-hinged veliger (D-shaped larva). (E) Eight-day-old, early, umbonal veliger showing indistinct umbo (u), extended velum (v) and digestive diverticuhnn (dd) as a dark patch covering viscerum. (P) Ten-day-old umbonal veliger. (G) 14-day-old umbonal veligers, including a larger, rapidly developing “streaker”. (H) 21-day old veligers with eye spots (es). (I) 24-day-old pediveliger with foot ( f) slightly extended. (.I) 25-day-old crawling plantigrade (metamorphosing pediveliger) showing extended foot and rudimentary gill filaments (gf) . (K) Recently settled, 28-day-old plantigrade (post-larva) showing dissoconch (d) and gill filaments. (L) 35day-old spat displaying gill filaments, partial extended foot, byssal threads (bt) and mantle (m), with dark pigment spots.

42

R.A. Rose, S.B. Baker/Aquaculture 126 (1994) 35-50

day 24. By day 25, plantigrades accounted for 22% of the population sampled, averaged 268 pm SL X 222 pm SH and ranged from 230 pm SL X 205 pm SH to 330 pm SL X 285 pm SH. At this time, they possessed rudimentary gill filaments and could be seen crawling with an extended foot (Fig. 1J) . Settlement occurred over 8 days (28-35) and on day 28, 5% of the population sampled were umbonal veligers, 60% pediveligers, 14% plantigrades and 23% spat, Newly settled spat averaged 349 pm SL X 289 pm SH with the smallest at 33 1 pm SL X 305 pm SH. They were attached to the substratum by clear byssal threads, and had rudimentary gill filaments and a finely reticulated, transparent dissoconch (adult shell) (Fig. 1K). By day 35, settlement had finished and spat which had settled on day 28 ( 1 week old) averaged 503 pm SL X 417 pm SH (ranged 410-820 pm SL), displayed a well-developed mantle with dark pigment spots, gill filaments and turquoise-coloured byssal threads (Fig. 1L). Scanning electron micrographs of the shell surface of 23-day-old veligers revealed that the prodissoconch I was smooth and that the prodissoconch II had growth rings (Fig. 2A and B). The provincular structure was simple (type a of Rees, 1950) and the number of taxodontal (rectangular) teeth at each end of the hinge on both valves increased with shell size. Four-day-old, straight-hinged veligers 90 pm SL, had 5 teeth per valve, with 3 teeth at the anterior end of the hinge line and 2 at the posterior end. Seven-day-old umbonal veligers 105 pm SL and 18-day-old veligers 180 pm in SL had 6 teeth per valve, with 3 at each end (Fig. 2C). Twenty-four-day-old pediveligers 230 pm SL had 9 teeth per valve with 4 at the posterior end and 5 at the anterior end (Fig. 2D). The transparent dissoconch of newly settled spat gradually turned opaque as the mantle became pigmented (Fig. 1K and J) . Spat less than 3-4 mm SL (or less than 3 weeks after settlement) were typically grey- or cream-coloured. As they developed, their mantles and shells began to display a variety of plain or zig-zag colour patterns which were white, green, orange, brown, red or purple-black and similar to those observed in the wild. Twenty-three or 24 days after settlement (50-51 days after fertilization), spat ranging from 2.8 mm SL X 1.8 mm SH to 4 mm SL X 3 mm SH developed a single growth process or finger on the left valve, approximately 42” to the left of the SH axis on the posterior side. A second finger formed on the left valve on day 30, or when spat averaged 4 mm SL X 3 mm SH, and by day 39 those which were 6 mm SL X 3.6 mm SH had a total of 4 or 5 fingers for both valves. Fingers developed initially along two or three ridges which radiated outwards from the umbo to margin of either valve. These ridges became obscure as growth processes grew and proliferated. Fingers of 4- and 5-month-old juveniles were supple, slightly convoluted and 2 or 3 times wider distally than proximally. At the posterior ear numerous fingers grew parallel and at right angles to the hinge (Fig. 3). The fingers of juveniles reared in downwellers for 5 months were similar to those transferred to sea after spending only 1 month in a downweller. ‘.Newly settled spat attached to the substratum with the right valve in contact with the bottom. Subsequently the right valve became slightly flatter and less convex than the left, which grew more rapidly, sometimes curving slightly over the lip of the right valve. The anterior ear of both valves was well developed (Fig. 3) and juveniles ranging from 10 mm SL X 6.6 mm SH to 15 mm SL X 10 mm SH were attached to the substratum by approximately 20 byssal threads. P. maxima spat were aggregate settlers and in some instances occurred in clusters of 2-8 individuals. On rare occasions individuals 3 or 4 weeks old left

R.A. Rose, S.B. Baker/Aquaculture

Fig. 3. Pinctada marima right: 2.5,3,4,4.3 and 5 paired valves of juveniles of individuals from same

126 (1994) 35-50

43

spat. (A) Outside view of the left valves of 5 juveniles at different ages, from left to months after fertilization, respectively. Note growth processes (gp). (B) Inside view of 5 months after fertilization (4 months after setting), illustrating different growth rates spawning. Note nacreous (n) and prismatic (p) layers of the shell. Scale in millimetres.

their original settlement site and were observed adhering to a long mucus strand and floating just below the surface in their culture vessels. The nacre on each valve was iridescent silver-white in colour (Fig. 3). The non-nacreous, prismatic layer was initially the same colour as the outside, periostracal layer and mantle, but by the time juveniles were 4 or 5 months old (5-6 months after fertilization) it had changed to brown.

R.A. Rose, S.B. Baker/Aquaculture

10

126 (1994) 35-50

15

20

25

30

Days after Fertilization

Fig. 4. Growth rate of the shell length (SL) of Pinctada maxima larvae up to 30 days after fertilization is described by the semi-logarithmic equation: SL = (76.271) ( 1.043)““, r=0.953, n= 345. Predicted curve is plotted as a solid line with 95% confidence limits shown as dash lines.

Growth and survival The mean water temperature was 26.5”C, 28.2”C and 29.2”C for the 3 batches cultured during 22 October 1987, 15 November 1988 and 28 November 1988, respectively. The growth rate from straight-hinged veliger to pediveliger stage not only varied between but also within each of the 3 batches, with increases in SL ranging from 6.1 to 14.2 pm-day - ’ and increases in SH from 5.8 to 12.1 pm*day-‘. The mean growth rates in SL over 24 days for larvae cultured at the lowest and highest mean temperature were 6.2 and 11.6 pmmday - *, respectively. When the growth data from all 3 batches were pooled, the mean growth rate of the larval shell over 24 days was 8 ,um*day-’ in length (Fig. 4) and 6.5 pm day-’ in height. These rates were similar to those of larvae from the second batch which were cultured at 28.2”C. The rate of growth along the length and height axes was described by the following semilogarithmic regressions (when day 0 was equal to the day of fertilization and n = 300) : SL= (76.271) ( 1.043)day, correlation

coefficient

(r) = 0.953 (Fig. 4)

and SH = (66.790) (1.044) &Y, I= 0.961. The mean growth rate from pediveliger to spat over 11 days varied less than 3 pm for all spawnings, and was 25 pm*day-’ in SL and 20 pm-day-’ in SH. There was no noticeable difference in the growth and survival of newly settled spat on the various types of collectors used. However, the mean growth rate of spat varied with the densities tested, regardless of settlement substratum. In both culture conditions (downwellers and open sea cages), the increase in SL averaged about 10 mm*mth-‘, with a range of 7.5-l 1.5 mm*mth-‘. Spat cultured in downwellers at a density of 4 individuals* 100 cm-’ averaged 9.6 mm*mtl-’ in SH while those stocked at 25 individuals* 100 cm-’ averaged 6 mm~mth’ (Fig. 5). When day 0 = fertilkation, the exponential equations describing growth in SH for spat cultured at the lower and higher densities, respectively, were:

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

45

SH = 4.32 X 10B5 day2.684, Y= 0.966, n = 390 and SH = 8.74 X 10e5 day2.445, r = 0.903, n = 1 223. Spat cultured at sea in circular, plastic trays at a density of 3 individuals* 100 cm-* averaged 9.2 mmsmth- ’ in SH while those stocked at 7 individuals* 100 cm-’ averaged 7.3 mm - mth - ‘. Faster growing spat and juveniles cultured in the hatchery or at sea increased in mean wet weight + 1 s.d. from 0.1 f 0.05 to 9.0 + 2.34 g (n = 20-22) in approximately 3 months (Fig. 6). Growth of the larval (prodissoconch II) and spat (dissoconch) shell was similar along the length and height axes. The prodissoconch II became oblong during the umbonal veliger stage (Fig. 11) and this shape was retained during formation of the dissoconch, with the shell broader anteriorly than posteriorly (Fig. 1K). The relationship between paired measurements of the SL and SH of 355 larval shells and 1241 spat shells randomly selected over the development period were described by growth equations (Gould, 1966). The exponential equation and correlation coefficient (r) for larval shells were H = 0.833L’.e**, z-= 0.993 and for spat they were H = 0.361L’~“g, r= 0.982. An analysis of covariance of the slopes of the logarithmic regressions indicated that they were not significantly different (P > 0.05). Although the common slope was estimated to be close to one ( 1.022, with a standard error of 0.00333), it was statistically greater than one (P~0.05) because of the large sample

Days after

Fertilization

Fig. 5. Growth rate of the shell height (SH) of Pinctada maxima spat reared at different stocking densities. Data plotted for a density of 4 spat- 100 cm* and 25 spat- 100 cm-’ are shown as (A) and (*), respectively. Predicted curve for a density of 4 spat - 100 cm* is described by the exponential equation: SH = 4.32 X 10-Sdayz.6s4, r= 0.966, n = 390. The curve and 95% confidence limits are plotted as dash lines. Predicted curve for a density of 25 spat/ 100 cm* is described by: SH= 8.74 X 10-5dayZ.445, r= 0.903, n = 1223. The curve and 95% confidence limits are plotted as solid lines.

46

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

1211. mean shell length

and height:

10. 9-

aS7E m 6. $

55 x 39mm .

6.

g 4. 3-

0

60

80

100

Days After

120

140

160

180

Fertilization

Fig. 6. Wet weight (g) of fast growing Pinctada marima spat reared at 25-32-C in the hatchery or at sea for 6 months. For various days after fertilization the weight is plotted as mean (n = 20-22)) standard deviation (rectangles) and range. Associated with each point plotted is the mean shell length and height (mm) of individuals sampled.

size. If a common slope was assumed, then there was a significant difference in elevation (P < 0.05) ; however, the actual difference (0.259 mm) appears to be slight. The total number of spat produced from 3 spawnings which were greater than 5 mm SL was estimated to be about 40 000 (i.e., less than 4% of the initial larval population). The percentage of straight-hinged veligers which successfully metamorphosed and settled, for each spawning, was: 1% for 22 October 1987; 1% for 15 November 1988; and 3% for 28 November 1988. The production rate (as defined by Alagarswami et al., 1989) ranged from 20 to 40 spatal-‘, with the higher rate of 40 spat*l-’ occurring in 500- and 1 000-litre culture tanks. Mortality of juveniles 6 months after settlement was l-2% for those reared in the hatchery and 9-12% for those reared at sea.

4. Discussion The general findings of this study were comparable with those of Tanaka and Kumeta ( 1981) for larvae and spat reared in Queensland. Both studies found the following developmental times and SL for various larval stages: straight-hinged veliger (day 1, 75-85 pm) ; umbonal veliger (days 8-12, 110-125 pm) ; and pediveliger (days 19-22, 180-230 ,um). The mean SL of the plantigrade larvae in both studies was approximately 270 pm, but the time at which the plantigrade stage occurred differs slightly. Those from Queensland appeared on days 19-2 1, while 80% of those from Western Australia appeared on days 24-

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

47

25. However, despite the variations in growth within and between batches from Western Australia, the mean growth rates of the larvae and spat from Queensland and Western Australian populations were similar, increasing 8 pm* day- ’ and 10 mm mth ’ in mean SL, respectively. The colour patterns, growth processes and general shape of hatcheryproduced spat were similar to those collected in the wild or reported in the literature by Tanaka and Kumeta (1981). A literature review by Alagarswami et al. ( 1989) revealed that the three species Pinctada ficata, P. margaritifera and P. maxima from the Indo-Pacific region have similar development. Despite variations in rearing conditions and natural differences between species and zoogeography, all three species are found to settle and metamorphose at approximately the same size (230-266 pm) and time (20-23 days after fertilization). Although the majority of larvae of P. maxima settled on days 24 and 25, this study supports the generalizations of Alagarswami et al. ( 1989) and further notes the following morphological discrepancies and similarities between the veligers and spat of each species. In P. fircata (Alagarswami et al., 1983) and P. margaritifera (Alagarswami et al., 1989)) the eye spots of the veligers are deeply pigmented and arise in individuals around 210 pm in SL. Eye spots in P. maxima are red-pigmented and generally occur in veligers 230 pm or greater in length, but occasionally they have been observed in veligers 200 pm in length (present study). The shell margin of P. margaritifera veligers has a pink-coloured tinge while that of P. fucata is colourless (Alagarswami et al., 1989). In P. maxima, the shell margin is light, yellow-brown or pink with the colour extending throughout the hinge structure as the veligers mature. The foot of the pediveliger and plantigrade larvae of P. maxima is translucent like that of P.fucata (Alagarswami et al., 1983) and not deeply pigmented like that of P. margaritifera (Alagarswami et al., 1989). Formation of the dissoconch in newly settled spat of P. maxima is practically identical to that described for the other two species. As in P. jkcata (Alagarswami et al. 1983), the valves of P. maxima grow unequally so that the left valve becomes more convex with age (Tanaka and Kumeta, 198 1). The growth processes on the shell of P. maxima are similar to that of P. murgaritifera, being broader distally than proximally, but each one is more curved than that of P. margaritifera. In P. ficata each process tapers into a fine point distally. As with P. ficata (Alagarswami et al., 1983), P. margaritijba (Alagarswami et al., 1989)) scallops and edible oysters (personal observations), the larvae of P. maxima exhibit variation in growth within a batch, with settlement typically lasting up to 1 week. This phenomenon is common to many bivalve species reared under hatchery conditions and may be associated with the following three husbandry practices used during this study. Firstly, the failure to rigorously cull slower-developing larvae from the population may have spread the settlement period over several days. When faster-developing larvae are selected for, the settlement period can be reduced to 3 days (Rose, unpublished data from 1990-1993 culture trials). Secondly, inappropriate preparation of collectors may have been partially responsible for an extended settlement period. Collectors used in this study were cured in seawater for only 2 weeks, which may have been too short a time for a bacterial film to adequately develop over the surface of the collectors and act as a stimulus to induce settlement and metamorphosis (Crisp, 1974). The delay in metamorphosis of bivalve larvae until a suitable stimulus or substratum is available has been well documented (e.g., Bayne, 1965; Crisp, 1974; Coon et al., 1990). Crassostrea gigas larvae, for example, have been shown experi-

48

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

mentally to delay metamorphosis in the absence of chemical stimulation but still retain their behavioural and morphogenetic competence for up to 30 days (Coon et al., 1990). Older larvae of C. gigas also have been experimentally induced to settle and metamorphose when exposed to supernatants of bacterial cultures (Fitt et al., 1990). Thirdly, a mixed algal diet was fed to P. maxima larvae as this has been shown to improve growth and metamorphosis in other bivalve larvae (Tan Tiu et al., 1989; Ferreiro et al., 1990). The diet selected, although high in fatty acids and lipids, may have inadvertently contributed to the extended settlement period since two of the four species used may have been unsuitable morphologically as food. One of these species, the chlorophyte, N. oculutu, has a thick cell wall which was not easily digestible and often passed intact through the digestive system of larvae. The other species, the diatom, C. gracilis, has elongated spines which hampered ingestion by smaller larvae. Thus, the larvae were not necessarily receiving an adequate supply of food. This may have resulted in under-feeding and, subsequently, a longer settlement period. Metamorphic competence in C. gigas larvae has been found to be correlated with, but not dependent on, size and eyespot development. Larvae of C. gigas can become competent before developing eyespots or fail to become competent even with well-developed eyespots (Coon et a1.,1990). P. maxima larvae during this study generally became morphologically competent at about 270 pm in shell length, with some attaining this state at sizes between 230 and 330 pm. Competent larvae always have a pair of fully developed pigmented eyespots, each approximately 10 pm in diameter. Minaur ( 1969)) however, has observed P. maxima larvae developing into pediveligers without eyespots and crawling. According to Minaur these larvae failed to attach because they appeared to be disturbed by the rapid proliferation of ciliates over the collector substrata. However, the lack of eyespots and the unusual settling behaviour could have been due to the gametes used by Minaur becoming damaged when they were forcibly matured with ammonium hydroxide. During this study, less than 4% of the initial larval population successfully settled and metamorphosed into spat which grew to 5 mm in shell length. A similar result of 6.3% was reported for P. margaritifera (Alagarswami et al., 1989). These low percentages suggest that the husbandry protocols used during both of these studies were less than optimal. Recent rearing trials with P. maxima have increased the percentage of larvae settling to 21-25% (Rose, 1991-1992, unpublished data). This translates into a production rate, as defined by Alagarswami et al. (1989)) of 188 spat-l-‘. The improvement appears to be largely due to feeding the larvae with phytoplankton which is morphologically more suitable for ingestion and digestion (i.e., Tahitian Zsochrysis and C. calcitrans) . Another reason may have been the practice of filling the settlement tanks with bundles of mesh collectors made from thick, old, frayed monofilament nylon netting. The additional surface area appeared to greatly reduce the incidence of aggregate settlement, a phenomenon which may have been responsible for reducing the percentage of larvae successfully settling during this study. Lightcoloured, semi-transparent rearing tanks used during this study may have inhibited settlement. Alagarswarmi et al. ( 1987) have found that dark-coloured tanks noticeably improve settlement over light-coloured ones for the larvae of P.fucutu. The various types of collectors used during this study did not conclusively demonstrate that one type was better than another. Glass or plastic plates appeared to collect more spat per unit area and to provide a better surface for the byssal attachment of each spat. However, reducing the density of spat on these collectors during nursery/grow-out culture was labo-

R.A. Rose, S.B. Baker/Aquaculture I26 (1994) 35-50

49

rious, requiring spat to be scraped off the surface and resettled onto new collectors. In contrast, the density of spat on monofilament mesh or shade cloth was reduced by simply teasing or cutting apart the collectors. Preliminary observations suggest that glass/plastic plates would appear to be a more suitable collector for spat cultured in areas of strong currents and large tides, whereas monofilament/shade cloth collectors would appear to be better suited for areas with less water movement. Survival rates of the spat reared in downwellers and in sea cages after 5 months was 9899% and 88-91%, respectively. Compared with levels reported for P. margaritijka of 1517% (Alagarswami et al., 1989), they were very high. Although these results are preliminary, they do suggest that the densities tested do not adversely affect survival. A commercial operation, however, may find that rearing spat in downwellers for 5 months is not cost-effective even though survival is high. Placing spat into grow-out 5-7 weeks after fertilization, when they are 2-4 mm in length, may be more cost-effective even though it is a more labour-intensive method. Tanaka and Kumeta ( 198 1) suggest that grow-out conditions of spat are enhanced by minimizing the effects of strong currents, large tidal cycles, and fluctuations in salinity. Except for little or no variation in salinity, these other factors did not appear to adversely affect spat growth or survival during this study. Approximately 20-25% of the spat produced at Broome were 120 mm SH 19 months after fertilization. This suggests that future commercial operations within the Broome area should plan on at least one out every 4 or 5 spat produced to be suitable for pearl culture at this time.

Acknowledgements The authors appreciate the cooperative efforts of the Western Australian Pearling Industry for providing wild broodstock. We are especially grateful to P. Fallon from the School of Veterinary Studies, Murdoch University, who prepared and photographed the larval shells with a scanning electron microscope. We also would like to thank K. Donohue and Dr. N. Caputi from the Western Australian Fisheries Department for their kind assistance with the statistical analysis of the growth data. This study was financed by the Australian Commonwealth Fishing Industry Research and Development Council.

References Alagarswami, K., Dharmaraj, S., Velayudhan, T.S., Chellam, A., Victor, A.C.C. and Gandhi, A.D., 1983. Larval rearing and production of spat of pearl oyster Pinctadaficata (Gould). Aquaculture, 34: 287-301. Alagarswarmi, K., Dharmaraj, S., Velayudhan, T.S. and Chellam, A., 1987. Hatchery Technology for pearl oyster production. Bull. Cent. Mar. Fish. Res. Inst., C.M.F.R.I., Co&in, India, 39: 62-71. Alagarswami, K., Dharmaraj, S., Chellam, A. and Velayudhan, T.S., 1989. Larval and juvenile rearing of blacklip pearl oyster, Pinctada margaritifera (Linnaeus) . Aquaculture, 76: 43-56. Bayne, B.L., 1965. Growth and the delay of metamorphosis of the larvae of Mytilus edulis (L.). Ophelia, 2( 1): 147. Chanley, M.H., 1981. Laboratory culture of marine bivalve molluscs. In: R.T. Hinegardner (Committee Chairman), Laboratory Animal Management: Marine Invertebrates. National Academy Press, Washington, DC, pp. 233-249.

50

R.A. Rose, S.B. Baker/Aquaculture

126 (1994) 35-50

Coon, S.L., Fitt, W.K. and Bonar, D.B., 1990. Competence and delay of metamorphosis in the Pacific oyster Crassostrea gigas. Mar. Biol., 106: 379-387. Crisp, D.J., 1974. Factors influencing the settlement of marine invertebrate larvae. In: P.T. Brant and A.M. Mackie, (Editors), Chemo-reception in Marine Organisms. Academic Press, London, pp. 177-265. Dybdahl, R.E. and Rose, R.A., 1986. The pearl oyster fishery in Western Australia. In: A.K. Haines, C.G. Williams and C. Coates (Editors), Tomes Strait Fisheries Seminar, Port Moresby, 11-14 February 1985. Aust. Gov. Publ. Service, Canberra. pp. 122-132. Ferreiro, M.J., Perez-Camacho, A., Labarta, U., Beiras, R., Planas, M. and Femandez-Reiriz, M.J., 1990. Changes in the biochemical composition of Ustrea edulis larvae fed on different food regimes. Mar. Biol., 106: 395401. Fitt, W.K., Coon, S.L., Walch, M., Weiner, R.M., Colwell, R.R. andBonar, D.B., 1990. Settlement behaviour and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supematants. Mar. Biol., 106: 389-394. Gould, S.J., 1966. Allometry and size in ontogeny and phylogeny. Biol. Rev.Pl: 587-640. Hynd, J.S., 1955. A revision of the Australian pearl shells, genus Pinctada (Lamellibranchia). Aust. J. Mar. Freshwater Res. 6: 98-137. Loosanoff, V.L., Davis, H.C. and Chanley, P.E., 1966. Dimensions and shapes of larvae of some marine bivalve mollusks. Malacologia, 4: 351435. Minaur, J., 1969. Experiments on the artificial rearing of the larvae of Pinctada maxima (Jameson) (Lamellibranchia). Aust. J. Mar. Freshwater Res., 20: 175-187. Pass, D.A., Dybdahl, R. and Mannion, M.M., 1987. Investigations into the causes of mortality ofthe pearl oyster, Pinctada maxima (Jameson). Aquaculture, 65: 149-169. Rees, C.B., 1950. The identification and classification of lamelibranch larvae. Hull. Bull. Mar. Ecol., 3( 19): 73104. Rose, R.A. and Dix, T.G., 1984. Larval and juvenile development of the doughboy scallop Chlamys (Chlamys) asperimus (Larmarck). Aust. J. Mar. Freshwater Res., 35: 315-323. Rose, R.A., Dybdahl, R., Sanders, S. and Baker, S., 1986. Studies on artificially propagating the gold or silverlipped pearl oyster, Pinctada maima (Jameson). In: R.E. Pyne (Editor), Darwin Aquaculture Workshop. Fisheries Div. Dept. Primary Indust. and Fish., Aust. Gov., Publ. Service, Canberra Tech. Rep. No. 3: 6067. Rose, R.A., Campbell, G.R. and Sanders, S.G., 1988. Larval development of the saucer scallop Am&urn balloti (Bemardi) (Mollusca: Pectinidae). Aust. J. Mar. Freshwater Res., 39: 153-160. Rose, R.A., Dybdahl, R.E. and Harders, S., 1990. Reproductive cycle of the Western Australian silverlip pearl oyster, Pinctadu marima (Jameson) (Mollusca, Pteriidae). J. Shellfish Res., 9( 2): 261-272. Tanaka, Y. and Kumeta, M., 1981. Successful artificial breeding of silver-lip pearl oyster, Pinctada maxima (Jameson). Bull. Natl. Res. Inst. Aquacult. Japan, 2: 21-28. Tan Tiu, A. Vaughan, D., Chiles, T. and Bird, K., 1989. Food value of eurytopic microalgae to bivalve learvae of Cyrtopleura costuta (Linnaeus, 1758), Crassostrea uirginica (Gmelin, 1791) and Mercenaria mercenaria (Linnaeus, 1758). J. ShellfishRes., 8(2): 399-405. Wada, S., 1953a. Biology and fisheries of the silver-lip pearl oyster. Library of C.S.I.R.O. Marine Laboratories, Hobart, Tasmania, Australia (unpubl.), 86 pp. Wada, S.K., 1953b. Biology of the silver-lip pearl oyster Pinctada maxima (Jameson). J. Artificial fertilization and development. Margarita, 1: 3-15.