Reproduction and spawning in calico scallops, Argopecten gibbus, from Bermuda.

ABSTRACT The gametogenic cycle and breeding season of the calico scallop, Argopecten gibbus was examined in Bermuda using quantitative and qualitative methods. Gonadal index varied seasonally, a rapid increase occurring in the winter months from January to March, and reaching maximum values in March and April. Greatest gonad growth and spawning were associated with low seawater temperature and low food levels. Variations in muscle indices implied a partial reliance on muscle reserves during the early stages of oogenesis. The later stages of ovarian growth (vitellogenesis and oocyte maturation) seemed to show a direct dependence on food supply. Histological analysis indicated the initiation of gamete development in early fall, when cell differentiation was first observed; oocyte ripening took place in late fall and early winter, reflected in increased oocyte diameters and gonadal indices. The lack of spawning activity during the hot summer months was seen histologically by the high proportion of spent cells in the gonads and a reduction in oocyte size, resulting in a minimal gonadal index. Individuals with gonad indices above 2 and mean oocyte diameters exceeding 50 [micro]m were induced to spawn with thermal shocks, exceeding 70% response rate in oocyte release in all trials. The present study extends our limited knowledge of the reproductive cycle in this subtropical scallop while providing a useful index to evaluate spawning readiness in cultured broodstock for routine hatchery purposes.

KEY WORDS: gametogenesis, scallops, spawning, reproduction, subtropical, Argopecten gibbus

INTRODUCTION

Calico scallops, Argopecten gibbus, are largely restricted to the subtemperate and tropical waters of western North Atlantic with the major stocks distributed from Cape Hatteras, North Carolina to the northeastern Gulf of Mexico (Waller 1969); calico scallops have also been collected from the Greater Antilles, Bermuda and the western portions of the Gulf of Mexico (Waller 1969). This species supports a commercial fishery off the East Coast of the United States (Blake & Moyer 1991), and its large-scale culture has been investigated in Bermuda for the purpose of stock enhancement and commercial activity since 1996 (Sarkis et al. 2003). Although much has been reported in the literature for the related bay scallop species, Argopecten irradians, little information is available for the calico scallop, especially for those inhabiting Bermuda waters. Data available from the commercial fishery in the United States has generated information on settlement, abundance and yield based on field studies (Roe et al. 1971), population assessments (Cummins 1971) and spawning periods in relation to environmental parameters (Allen 1979, Miller et al. 1981).

Knowledge of the reproductive cycle is necessary for restocking and/or mariculture purposes. Moyer and Blake (1986) have assessed the reproductive cycle of calico scallops in Florida using gonadal indices. However, as documented in Argopecten irradians, latitudinal differences occur in the timing of gametogenesis and reproduction; in general, reproductive events occur later in the year and at higher water temperatures in more southern latitudes (Barber & Blake 1991).

The present study describes the gametogenic cycle of Argopecten gibbus inhabiting Bermuda waters. Although the use of gonadal indices is a simple, fast and inexpensive method, which may be routinely used in hatcheries, verification of reproductive events pertaining to gamete development is obtained by histological techniques (Couturier & Newkirk 1991). In the present study, quantitative (gonad index) and qualitative (histology) methods were used to assess the reproductive status of calico scallops. This further allowed for testing the validity of relying on gonadal indices for controlled spawning induction of A. gibbus in the hatchery.

MATERIALS AND METHODS

Calico scallops reared in suspended cultures in Harrington Sound, Bermuda (32[degrees]N, 64[degrees]W) were collected on a monthly basis from December 2000 to December 2002. Each sample consisting of 30 scallops >50 mm in shell height was transported in water to the pilot hatchery at the Bermuda Biological Station for Research (BBSR) and maintained in an open filtered seawater system for a 24-h period, allowing for gut clearance. Scallops were scrubbed of epibiota and shell height and length measured to the nearest 0.1 mm, with vernier calipers. Scallops were blotted dry and total wet weight was determined using a Sartorius balance (to nearest 0.01 g).

A subsample of 15 scallops was used for determination of gonad and muscle indices on a monthly basis for a two-year period from December 2000 to December 2002; in the second year of sampling, December 2001 to December 2002, a second monthly subsample of 10 scallops was taken simultaneously for histological analyses.

Tissue Indices

Various condition indices have been used to characterize the apparent "health" of bivalves. Lucas & Beninger (1985) in a critical review of several of these indices, concluded that the dry tissue weight:dry shell weight ratio gave meaningful information about the physiological state of the animal, including reproductive activity. In the present study, gonad and muscle indices were calculated separately; the former reflecting gonad development, and the latter as an indication of seasonal metabolic changes, possibly associated with gametogenesis (Sarkis 1993). Gonads and adductor muscles were dissected, and dried to constant weight in an oven at 80[degrees]C. Tissue indices were calculated as follows: (Dry weight of tissue/dry empty shell weight) x 100; where a high gonadal index is indicative of mature gonads, and a lower index reflects the onset of gametogenesis or spent gonads.

Histology

To obtain further insight into oogenesis and oocyte maturation, samples of gonads were dissected monthly and prepared for histological analysis. A small distal portion of the gonad (5 mm x 5 mm) was fixed in Bouin's fluid for 48-72 h and stored in 70% ethanol until sectioning (Naidu 1970). Fixed ovarian tissue samples were processed for histology according to standard methods for marine bivalves (Howard et al. 2004). The fixed tissues were dehydrated through an ascending alcohol series, embedded in wax, cut to 5-[micro]m sections and stained with hematoxylin and eosin.

Thin sections were examined under a compound microscope (x100 magnification) for developmental purposes and categorized into 6 stages based on Naidu (1970) and Couturier and Newkirk (1991) as follows--Stage 1: undifferentiated, follicular structure is indistinct, and no sex cell primordia; Stage 2: differentiated, oocyte development commences, primary oogonia are visible, sex cells range from 15-25 [micro]m in diameter; Stage 3: developing, mostly previtellogenic and vitellogenic oocytes, follicle lumen is open or only partially occluded, and oocyte diameter ranges from 30-70 [micro]m; Stage 4: ripe, gonads show packed follicles, the fob licular lumen is occluded, there is no interfollicular space visible, and oocytes range 60-70 [micro]m in diameter; Stage 5: spawning, the follicle lumen is opening, follicle size diminishing and ripe oocytes range 60-70 [micro]m; Stage 6: spent, follicles are empty and frequently filled with hemocyte aggregations on slide preparations, and an occasional atretic oocyte is present.

Each scallop was categorized according to one of the previous gonad development stages, and the frequency (%) of each stage determined for monthly samples. The data were plotted as cumulative histograms while assessing the developmental stage of each scallop under the microscope, the diameters of 10 oocytes were measured randomly in each individual using an ocular micrometer. Photomicrographs of representative developmental stages were obtained under x100 magnification using a digital photomicroscopy setup (see Fig. 3 later).

[FIGURE 3 OMITTED]

Spawning Induction

Results from spawning induction, performed during the course of four years of hatchery operation (1996-1998, 2002), are presented here for comparison with conclusions drawn from in-depth histological data determined within the scope of this study. Cultured adults were obtained from field sites during winter/early spring (December to April), and the gonadal index calculated as previously described on a subsample. Spawning induction was attempted on 30 scallops by standard thermal cycling shocks (Sarkis et al. 2003, Bourne et al. 1989). The proportion of spawning females was calculated for each trial, as an indication of spawning success. Comparisons were made between the percentage response and histological state.

RESULTS

Tissue Indices

Gonadal and muscle indices for the calico scallop reared in Bermuda waters are shown in Figures 1 and 2, respectively. Although fluctuations in gonadal indices were large, especially in the first year of sampling, the trend was for maximum values of the index during the winter months, and minimum values during the summer months. There was a slight increase in gonadal indices during the autumn months (September to December), remaining relatively low until December (0.2-0.9). Thereafter, a rapid increase in gonad weight occurred in the winter months, from January to March, reaching maximal values in March and April.

[FIGURES 1-2 OMITTED]

The time required for scallops to attain maximum gonad index varied slightly from year to year, with maximum values observed in April in 2001 (3.07 [+ or -] 1.18), and in March during 2002 (2.13 [+ or -] 0.83). Variation in gonad condition among individual scallops was large particularly throughout the spawning season from February to May, implying perhaps low synchronicity in spawning among conspecifics. Gonadal indices were at their lowest during the summer months from June to August, ranging from 0.12 ([+ or -] 0.38) to 0.17 ([+ or -] 0.12) in 2001, and 0.33 ([+ or -] 0.14) to 0.41 ([+ or -] 0.12) in 2002, representing a "rest" period.

The trend in adductor muscle indices was not as distinct as that seen for gonadal indices and seasonal fluctuations were not as wide, although they were apparent (Fig. 2). Maximum values were generally recorded at the end of the summer months in September, reaching 5.33 [+ or -] 0.65 in 2001 and 4.99 [+ or -] 0.92 in 2002. These coincided with low gonadal indices (compare Figs. 1 and 2). Muscle indices decreased throughout the autumn, as gonadal indices started to increase. As gonadal indices peaked from January to April, muscle indices remained low, fluctuating between 3.01 [+ or -] 0.47 and 4.06 [+ or -] 1.03 in both years of the study. Variations among individuals were seen on a monthly basis, as for gonadal indices.

In summary, gonadal indices indicated large seasonal fluctuations with the occurrence of well-defined peaks of gonadal activity during the winter months (January to April), and a period of inactivity during the summer months. Muscle indices showed, to a certain extent, inverse seasonal fluctuations to those recorded for gonadal indices.

Histology

Photomicrographs of ovarian developmental stages are shown in Figure 3. Each stage is easily identified by the type of cells present or absent, the shape of the follicles containing gametes, and the size of the oocytes.

The first appearance of cell differentiation was observed in September, when the gonad consisted of 60% developing cells (Fig. 4). The first appearance of ripe oocytes was recorded in December, and the gonad at this time consisted mainly of developing and ripe cells (90% in 2001 and 100% in 2002) (Fig. 4). Oocyte ripening occurred during the winter months, from December to March, as the proportion of ripe cells increased to 80%, 90% and 70% of the gonad in February, March and April, respectively (Fig. 4). During these three months, the gonad samples consisted of 100% ripe and spawning individuals, implying a readiness to spawn. Between April and June, the proportion of spawning and spent individuals increased, indicating the end of spawning activity (Fig. 4). By the month of August, 100% of the scallops seen were spent and contained no gametes (Fig. 4).

[FIGURE 4 OMITTED]

Seasonal variations in mean oocyte diameter are recorded in Figure 5. Microscopic observation of histological sections of ovaries revealed that oocyte size was comparable in individuals collected during oocyte ripening in the winter, from December to April, ranging from 53.7 [micro]m to 63.5 [micro]m. The dramatic decrease in oocyte size to 23.1 [micro]m seen from April to June coincided with the histological evidence of spawning and the lack of gametogenic activity seen during the hot summer months.

[FIGURE 5 OMITTED]

Spawning Induction

Results obtained in the hatchery from spawning induction trials yielded a spawning success of 70% to 90% of the females induced (Fig. 6) when gonadal indices exceeded 2. Such individuals generally responded to the thermal treatment and released gametes within 90 min of stimulation and most individuals did so in a synchronous manner. Attempts to spawn individuals with gonadal indices below or equal to 2 resulted in variable success with generally fewer than 40% of individuals responding to the thermal shock (Fig. 6).

[FIGURE 6 OMITTED]

DISCUSSION

The present study suggests that calico scallops in Bermuda undergo reproduction mainly from December to April, and possibly May. Gamete differentiation and development began in September, associated with a mean oocyte size of <20 [micro]m, low gonadal indices and maximal muscle indices. Oocyte development occurred throughout the fall, coinciding with a gradual increase in mean oocyte diameter and gonad weight, and a reduction in adductor muscle indices. The first ripe scallops appeared in December when mean oocyte diameter exceeded 50 [micro]m and gonadal index approached 1. Oocyte ripening continued during the winter months, associated with low muscle indices and a rapid increase in gonad weight yielding by February a gonadal index close to or above 2 with 80% of the scallops classified as ripe and showing some evidence of spawning in histological sections. The spawning period extended into March and April, and was reflected in the responses obtained from controlled spawning trials at the BBSR hatchery, where induced individuals released gametes between February and April. Optimal spawning responses were obtained when gonadal index was above 2, coinciding with histological analyses showing ovaries with high percentage of ripe oocytes (>70%). Alter the spawning period, both gonadal index and oocyte diameters declined rapidly in late spring/early summer, and histology showed increased proportions of spent individuals. By August, gonads consisted of spent cells only, resulting in minimal gonadal indices and coinciding with a lack of spawning activity during the summer months.

The spawning period of the calico scallop in Bermuda, extending over the winter months, is associated with colder water temperatures and lower food availability. This cycle differs from that recorded for the same species in Florida (Moyer & Blake 1986), where two spawning periods are observed, the first in late spring (April to June), and the second in the autumn. As mentioned previously, latitudinal differences in gametogenesis have been recorded within a species, explained in part by spatial differences in environmental influences, namely those of food and temperature (Barber & Blake 1985, MacDonald & Thompson 1988).

Temperature is most often cited as influencing bivalve reproduction and triggering spawning. Gamete maturation for Argopecten gibbus in Bermuda coincided with minimal sea surface temperatures ranging from 16[degrees]C to 18[degrees]C (Doug Connelly, unpublished data). Blake & Moyer (1991) found that the required threshold temperature for gonad development and ripening for the same species in Florida was of 19[degrees]C to 20[degrees]C. They further concluded that temperatures above 22[degrees]C stopped gonad ripening and spawning did not occur, which agrees with the results in the present study where spawning terminated in late spring at temperatures above 21[degrees]C, and a lack of gametogenic activity was seen during the warmer months (July and August), when sea surface temperatures reached a maximum (29[degrees]C) (Doug Connelly, unpublished). However, in keeping with the annual variations in gonadal indices reported in the present study, Manuel (2001) recorded yearly fluctuations in peak spat settlement periods in the field. These variations may be associated with changes in sea surface temperatures and food supply between years, because environmental characteristics of Bermuda's inshore waters, habitat to these scallop species, have been known to be wind-driven and hence highly variable; this is especially apparent in the timing of "bloom" conditions, expressed as chlorophyll a (1.93 [micro]g chl a.[L.sup.-1]), occurring during late fall or early winter (November and December) (Doug Connelly, unpublished). The trend seen in the muscle index points to the relationship between "bloom" occurrences and gonadal development in the early stages of oogenesis, and it implies to a certain extent a direct dependence on external food supply for oocyte ripening during periods of low food availability in late winter/early spring (0.42 [micro]g chl a.[L.sup.-1]) (Doug Connelly, unpublished). Similar conclusions were drawn by Manuel (2001) for another scallop species inhabiting Bermuda waters, Euvola (Pecten) ziczac.

Thermostimulation has proven efficient and rapid for inducing spawning in calico scallops in Bermuda with a mean gonadal index above 2, compared with other species (Monsalvo-Spencer et al. 1997). To date, thermal shock seems to be the most efficient method, for inducing sperm and ova release in hatchery scallops (Monsalvo-Spencer et al. 1997), and this is especially true for pectinidae, where a critical temperature range or minimum threshold temperature has been most frequently implicated in the initiation of spawning (Barber & Blake 1983, Moyer & Blake 1986). However, variable responses have been obtained using thermal shock induction in Bermuda with scallops having a gonadal index close to or below 2. This points to the influence of other factors in controlled spawning events, perhaps related to hatchery procedures and/or to natural spawning cues. For example, it has been shown that spawning in cultured and natural populations of various scallops does not appear to occur until the majority of the animals in a population have ripe gonads, normally above 70% or more of the populations (Lubet et al. 1991, Couturier 1994, Lohrmann & Von Brand 2005). In this study, fluctuations seen in gonadal index among individuals on a monthly basis indicate a heterogeneity within the sample, and should be given closer consideration when mean index is low; it may well be that sample size for hatchery spawning induction needs to be increased at this time, to ensure a larger proportion of ripe animals. In addition, it may be worthwhile investigating the influence of other spawning cues on calico scallops in Bermuda. Several of these, such as salinity, lunar phase, light, dissolved oxygen, pH, mechanical shock and various chemicals and food pulses have been examined for other pectinid species through correlation with field observations and more rigorous laboratory experiments (Gibbons & Castagna 1984, Starr et al. 1990, Velez et al. 1990, Desrosiers & Dube 1993, Jamaillo et al. 1993, Couturier 1994).

In conclusion, it seems necessary for the calico scallop to have a mean gonadal index value >2 for a good response to spawning induction via thermal shocks. It can be further added that the use of gonadal indices for routine hatchery purposes provides a good snapshot of the reproductive status. However, to have a thorough understanding of gametogenic events in a specific population in relation to gonadal indices, histological assessments of reproduction in scallops is warranted in a first instance.

LITERATURE CITED

Allen, D. M. 1979. Biological aspects of the calico scallop, Argopecten gibbus, determined by spat monitoring. Nautilus 94:107-119.

Barber, B. & N. J. Blake. 1983. Growth and reproduction of the bay scallop, Argopecten irradians (Lamarck) at its southern distributional limit. J. Exp. Mar. Biol. Ecol. 66:247-256.

Barber, B. J. & N. J. Blake. 1985. Intra-organ biochemical transformations associated with oogenesis in the bay scallop, Argopecten irradians concentricus (Say), as indicated by 14C incorporation. Biol. Bull. 168(2):39-49.

Barber, B. J. & N. J. Blake. 1991. Reproductive physiology. In: S. Shumway, editor. Scallops: biology, ecology and aquaculture. Developments in Aquaculture and Fisheries Science, vol. 21, Amsterdam, The Netherlands: Elsevier Science Publishers B.V. pp. 377-420.

Blake, N. J. & M. A. Moyer. 1991. The calico scallop, Argopecten gibbus, fishery of Cape Canaveral. Florida. In: S. Shumway, editor. Scallops: biology, ecology, and aquaculture. New York: Elsevier. pp. 899-909.

Bourne, N., C. A. Hodgson & J. N. C. Whyte. 1989. A manual for scallop culture in British Columbia. Can. Tech. Rep. Fish. Aquat. Sci. No. 1694. 215 pp.

Connelly, D. unpublished. Bermuda inshore water investigation. Bermuda Biological Station fur Research Inc., 1988-2002.

Couturier, C. 1994. Spawning in sea scallops, Placopecten magellanicus. In: N. F. Bourne, B. L. Bunting & L. D. Townsend, editors. Proceedings of the 9th International Pectinid Workshop, Nanaimo, BC, Canada, vol. 1. Can. Tech. Rep. Fish. Aquat. Sci. pp. 138-146.

Couturier, C. & G. Newkirk. 1991. Biochemical and gametogenic cycles in scallops, Placopecten magellanicus (Gmelin, 1791), held in suspension culture. In: S. Shumway & P. Sandifer, editors. An international compendium of scallop culture and biology. World Aquaculture Workshops, No. 1, World Aquaculture Society Baton Rouge, LA. pp. 107-117.

Cummins, R. 1971. Calico scallops of the Southeastern United States, 1959 to 1969. NOAA Technical Report NMFS SSRF-657. 22 pp.

Desrosiers, R. R. & F. Dube. 1993. Flowing seawater as an inducer of spawning in the sea scallop, Placopecten magellanicus (Gmelin, 1791). J. Shellfish Res. 12(2):263-265.

Gibbons, M. C. & M. Castagna. 1984. Serotonin as an inducer of spawning in six bivalve species. Aquaculture 40:189-191.

Howard, D. W.. E. J. Lewis, B. J. Keller & C. S. Smith. 2004. Histological techniques for marine bivalve mollusks and crustaceans. NOAA Technical Memorandum NOS NCCOS 5. 218 pp.

Jamaillo, R., J. Winter, J. Valencia & A. Riviera. 1993. Gametogenic cycle of the Chiloe scallop (Chlamys amandi). J. Shellfish Res. 12(1):59-64.

Lohrmann, K. B. & E. Von Brand. 2005. Histological study of gonads of triploid scallops, Argopecten purpuratus. J. Shellfish Res. 24(2):369-375.

Lubet, P., R. Faveris, J. Y. Besnard, I. Robbins & P. Duval. 1991. Annual reproductive cycle and recruitment of the scallop Pecten maximus (Linnaeus, 1758) from the Bay of Seine. In: S. Shumway & P. Sandifer, editors. An international compendium of scallop culture and biology, World Aquaculture Workshops, No. 1. World Aquaculture Society, Baton Rouge, LA. pp. 87-94.

Lucas, A. & P. G. Beninger. 1985. The use of physiological condition indices in marine bivalve aquaculture. Aquaculture 44:187-200.

MacDonald, B. A. & R. J. Thompson. 1988. Intraspecific variation in growth and reproduction in latitudinally differentiated populations of the giant scallop Plaeopecten magellanicus (Gmelin). Biol. Bull. 175(3):361-371.

Manuel, S. 2001. Reproduction and spat settlement of Euvola ziczac around Bermuda. Ph.D. thesis. University of Liverpool, UK.

Miller, G. C., D. M. Allen & T. J. Costello. 1981. Spawning of the calico scallop Argopecten gibbus in relation to season and temperature. J. Shellfish Res. 1(l):17-21.

Monsalvo-Spencer, P., A. N. Maeda-Martinez & T. Reynoso-Granados. 1997. Reproductive maturity and spawning induction in the catarina scallop Argopecten ventricosus (= circularis) (Sowerby II, 1842). J. Shellfish Res. 16(1):67-70.

Moyer, M. & N. J. Blake. 1986. Fluctuations in calico scallop production (Argopecten gibbus). Proceedings of the Eleventh Annual Tropical and Subtropical Fisheries Conference of the Americas. pp. 45-58.

Naidu, K. S. 1970. Reproduction and breeding cycle in the giant scallop. Placopecten magellanicus (Gmelin) in Port au Port Bay, Newfoundland. Can. J. Zool. 48:1003-1012.

Roe, R. B., R. Cummins, Jr. & H. R. Bullis, Jr. 1971. Calico scallop distribution, abundance, and yield off eastern Florida, 1967-1968. Fish. Bull. (Wash. DC.) 69:399-409.

Sarkis, S. 1993. Seasonal changes in the gross biochemical composition of the turkey-wing, Arca zebra (Swainson, 1883), in Bermuda. J. Shellfish Res. 12:329-336.

Sarkis, S., M. Helm, A. Cogswell & P. Farrington. 2003. Calico scallop culture in Bermuda: a low cost, pilot hatchery for the tropics and subtropics. In: R. LeRoy Creswell, editor. Proceedings of the Fifty-Fourth Annual Gulf and Caribbean Fisheries Institute., Gulf and Caribbean Fisheries Institute Inc., Fort Pierce, Florida. pp. 488-495.

Stare M., J. H. Himmelman & J.-C. Therriault. 1990. Direct coupling of marine invertebrate spawning with phytoplankton blooms. Science 247:1071-1074.

Velez, A., A. Alifa & O. Azuaje. 1990. Induction of spawning by temperature and serotonin in the hermaphroditic tropical scallop Pecten ziczac. Aquaculture 84:307-313.

Waller, T. R. 1969. The evolution of the Argopecten gibbus stock (Mollusca:Bivalvia) with emphasis on the tertiary and quarternary species of eastern North America. J. Paleont. 43(Suppl.): 1-125.

SAMIA SARKIS, (1)* CYR COUTURIER (2) AND ANDREW COGSWELL (3)

* Corresponding author. E-mail: [e-mail]

(1) Department of Conservation Services, P.O. Box FL 588, Flatts FLBX, Bermuda; (2) Marine Institute, Box 4920, St. John's, NL, Canada, A1C 5R3; (3) Bermuda Biological Station for Research, Ferry Reach GEO1, Bermuda