Heterosis between two stocks of the bay scallop, Argopecten irradians irradians Lamarck (1819).

ABSTRACT Heterosis between two stocks (A and B) of the bay scallop Argopecten irradians irradians (Lamarck, 1819) was studied by comparing performance in hatching success, larval survival, larval, spat and adult growth of reciprocal hybrid crosses (AB and BA) and that of two parental groups (AA and BB). Stock A is derived from the initial 1982 introduction from Connecticut, USA and expected to be highly inbred because of the small founder population (26 individuals) and the long breeding history (20 + generations). Stock B is the fourth generation from a 1999 introduction from Virginia and Massachusetts, USA, which is expected to have higher heterozygosity caused by a larger founder population (200 wild and 206 first generation cultured individuals) and a short breeding history. Both hybrid crosses demonstrated positive heterosis for all traits studied, although the magnitude of heterosis varied among traits and life history stages. Midparent heterosis was 7.9% for hatching success, 10.8% and 3.0% for larval survival and growth and 8.4% and 10.9% for juvenile and adult growth. Single-parent heterosis also varied among traits. For fitness related traits such as hatching success and larval survival, single-parent heterosis was positive for both parental stocks. Single-parent heterosis for growth, a morphological trait, was positive in stock A and negative in stock B.

KEY WORDS: heterosis, scallop, Argopecten irradians irradians, stocks, pair-crosses

INTRODUCTION

Observations of heterosis or hybrid vigor (Shull 1914) are centuries old. Shull (1952) defined the term heterosis as "the interpretation of increased vigor, size, fruitfulness, speed of development, resistance to disease and to insect pests, or to climatic rigors of any kind, manifested by crossbred organisms as compared with corresponding inbreds, as the specific results of unlikeness in the constitutions of the uniting parental gametes." This definition, however, is often interpreted as not implying a genetic basis for heterosis, because the definition basically describes the phenotype that results from crossing two different inbred lines (Lamkey & Edwards 1998). Therefore, the definition of heterosis or hybrid vigor as the difference between the F1's and the mean of the two parents (Falconer 1981) is commonly used in the literature, and this heterosis is often called midparent heterosis.

Despite its tremendous success in plant and animal breeding, the genetic basis of heterosis remains uncertain (Griffing 1990, Luo et al. 2001). Three hypotheses of nonadditive gene action may explain heterosis: overdominance (Shull 1908, East 1936), dominance (Bruce 1910) and epistasis (Stuber et al. 1973, Wright 1977). In contrast with overdominance, neither dominance nor epistasis requires heterozygote superiority at any locus (Hedgecock et al. 1996). In spite of the tremendous success of utilizing heterosis in crop improvement (Crow 1998), its exploitation in marine bivalves has been limited and more recent. Heterosis in bivalves was first suggested by Singh and Zouros (1978) based on a positive correlation between allozyme heterozygosity and fitness-related traits in individuals from natural populations. Mallet and Haley (1983) first demonstrated the presence of heterosis in marine bivalves using experimental crosses between different populations of the American oyster Crassostrea virginica. Hedgecock et al. (1995) measured heterosis for quantitative traits using crosses among inbred lines of Pacific oyster Crassostrea gigas. In recent year, quantitative studies of heterosis using experimental crosses have been conducted in several marine bivalves (Hedgecock et al. 1996, Cruz & Ibarra 1997, Cruz et al. 1998, Beaumont et al. 2004).

The bay scallop Argopecten irradians, a marine bivalve of considerable economic importance, is a functional hermaphrodite that simultaneously spawns eggs and sperm for external fertilization. Like other functional hermaphrodites, self-fertilization is common and can result in rapid inbreeding (Stiles & Choromanski 1995, Zhang et al. 2003). In 1982, the northern subspecies, A. i. irradians, was first introduced into China successfully from the United States and developed into one of the most important mariculture industries in China (Zhang et al. 1986). By early 1990, the annual production of bay scallops in China had reached about 200,000 tons (Guo et al. 1999). For many years, however, bay scallop production in China was from the 26 founders introduced in 1982 and as a result, a greater proportion of mtDNA variation was lost in the intervening generations of hatchery breeding (Blake et al. 1997). Additional scallops (406 including 200 of wild and 206 from the first generation of a captivity population) were brought from the United States to China in December 1998 and February 1999 (Gu Z., personal communication), which currently are the main culture stock in China.

In China, bay scallop aquaculture has matured to include a series of well-defined phases such as broodstock conditioning, larval culture, nursery and grow-out. Broodstock conditioning and larval culture are conducted in hatcheries under strict control. Nursery and growth use bags and lantern nets respectively, and adults are harvested before sexual maturation. Although the scallops deriving from different stocks are often cultured in the same sea area, no mixing of the stocks has occurred. Also, natural populations of bay scallop have been not found in China (personal observation). Therefore, gene flow between different stocks is believed to be limited. The two stocks of bay scallop in China may represent different genetic stocks. When genetic improvement programs are planned, the design should be based not only on the distinction of available stocks, but also on the genetic characteristics of important production traits at different life stages during hatchery and field grow-out phases (Cruz & Ibarra 1997). In China, the bay scallop has been exclusively produced in hatcheries for more than two decades, and hatchery production of seed provides an opportunity for genetic improvement.

To determine if heterosis exists and can be used for genetic improvement in the bay scallop, four groups including two reciprocal hybrid crosses and their parental controls were produced by pair-mating between two stocks and mass-spawning within each stock. The four groups were reared under the same conditions to minimize effects of environmental factors. Heterosis was estimated and compared for several traits at different life stages, including hatching success of fertilized-eggs, larval survival, and growth at larvae, spat and adult stages.

MATERIAL AND METHODS

Parental Stocks and Conditioning

The two parental stocks of bay scallops used in this study, named "A" and "B", were the same age and had been cultured under the same environmental conditions in the Laizhou Bay of Bohai Sea, but they differed in their origin and breeding history. Stock A is derived from the first introduction from Connecticut, USA in 1982 and has been successively cultured in China for more than 20 generations. Stock B is the forth generation from a recent introduction from Virginia and Massachusetts USA in 1999. The two parental stocks used in the present study were produced as the control lines (nonselected) for a selection experiment in 2002 (Zheng et al. 2004a). In March 2003, shell lengths of a random sample of 200 individuals from each stock were measured with a vernier caliper (accuracy, 0.02 mm), and 500 individuals of the two stocks were conditioned using the method described by Zheng et al. (2004a). In short, scallops of two stocks were placed separately in lantern nets (8 layers/net and 20 scallops/layer) and conditioned in a 50 [m.sup.3] concrete tank. Each scallop was fed with 4 x [10.sup.6] cells of Nitzschia closterium daily. Water temperature was raised from 7-8[degrees]C to 16[degrees]C, and kept at 16[degrees]C for a week. Next, temperature was gradually raised from 16[degrees]C to 18[degrees]C, and kept at 18[degrees]C thereafter. Salinity ranged from 30-31 ppt. Water was changed and feces removed by siphoning from the bottom of the tank daily. After approximately one week at 18[degrees]C, most of the breeders were ripe or reached ripeness stage IV, as visually determined according to Sastry (1963).

Experimental Design and Treatments

On May 4, 10 sexually mature scallops with gonadal condition at stage IV (Sastry 1963) were chosen from each stock for spawning. To induce spawning, the combined method of by injecting serotonin into scallop's adductor muscle (Cruz & Ibarra 1997) and thermal shock by raising water temperature from 18[degrees]C to 23[degrees]C was used, and the following procedure was able to prevent self-fertilization. First, each scallop was injected with 0.1 mL of 0.02 mM serotonin (5-hydroxytryptamine, Sigma) into the adductor muscle. Next, each spawner was separately placed in one 5-1 polyethylene bucket containing 3-1 of 23[degrees]C seawater (filtered to 30 [micro]m). After 15-20 min, all animals began to release sperms. Each spawner released sperm several times during a 1-h period. Then, each spawner was taken out and rinsed 2-3 times with 23[degrees]C seawater and individually placed in one 5-L bucket containing 3-L filtered seawater at 23[degrees]C. In general, most individuals began to release eggs after about 20 min. Each spawner released eggs in 2-3 expulsions during approximately 30 min. After spawning, eggs were washed on a 30-[micro]m screen to remove sperm if any and placed into one 40-L bucket, respectively. Fecundity was estimated by counting the number of eggs from each scallop. The diameter of 20 eggs from each spawner was measured under microscope (xl00). Eggs were sampled and double-checked under microscope for fertilization caused by sperm contamination, and eggs were discarded if they were contaminated.

Eggs of each female were divided in two parts, one was pair-crossed to a male from the other stock, and the other part was pooled first within one stock and then mass-crossed with mixed sperm from the same stock. Therefore, four distinct groups were made and used in the study, including two reciprocal hybrid crosses AB (AS x B[female]) and BA (B[male] x A[female]) produced by pair-crossing between two stocks and two parental groups AA (A [female] x A [male]) and BB (B[female] x B [male]) produced by mass-crossing within each stock. Fertilized eggs were placed in polyethylene buckets for incubation. Twenty reciprocal-crosses were placed respectively in twenty 40-L buckets, and each parental group was separated into three replicates, and then each replicate was placed in one 80-L bucket. Water temperature and salinity during incubation were 23[degrees]C and 30 ppt, respectively.

Thirty hours after fertilization, D-stage larvae hatched, and hatching success of fertilized-egg was estimated by counting the number of normal D-stage larvae per replicate. Afterward, the larvae from the same reciprocal cross group were collected from each bucket using a 30-[micro]m sieve, placed mixed in one 80-L polyethylene bucket with seawater and divided into three parts equally, conforming 3 replicates as with the parental stocks. Initial stocking density was 10 larvae [mL.sup.-1] for all groups.

To minimize environmental effects, all four groups were reared under the same condition during larval culture, spat nursery and adult grow-out according to methods described by Zheng et al. (2004a). No culling of small individuals was conducted at any stage, therefore reducing potential effects of altering the evaluated traits.

Trait Assays

Four traits were measured: (1) parental fecundity, measured as the average number of eggs spawned by each spawner; (2) hatching success, measured as the ratio of the number of normal D-stage larvae to the number of fertilized eggs; (3) larval survival, as the ratio of the numbers of viable larvae at culture days 4, 7, and 10 to the numbers of larvae on day 1; and (4) growth, measured as shell length at larval, spat, and adult life stages.

The numbers of eggs, normal larvae at day l, and larvae at days 4, 7 and 10 were estimated as the average of three samples of 1 mL each using a dissecting microscope (x20), extrapolating numbers then to total volume. Shell length (the longest anterior-posterior distance) of 30 randomly taken individuals per replicate was measured using different methods, depending on the life stage. Larvae were killed with 4% formaldehyde, and measured on days 1, 4, 7 and 10 after using a microscope (xl00) equipped with an ocular micrometer. Spats were measured alive on days 20, 30, 40 and 50 using a dissecting microscope (x20-40) equipped with an ocular micrometer. Adults were also measured alive at days 70, 100, 130 and 160 using a Vernier caliper (accuracy: 0.02 mm).

Estimate of Heterosis

Because scallops in commercial hatcheries have been consistently produced by the method of mass spawning, the two mass crossed groups (AA and BB) can be considered as a representative sample of the two populations to estimate heterosis in this study.

In this study, midparent heterosis ([H.sub.MP]) is defined as the difference between the mean of the reciprocal hybrid crosses and the mean of the two parents (Falconer 1981) and calculated by the following equation:

[H.sub.MP]% = FC - P/P x 100

where, FC = average phenotypic value of the reciprocal crosses, P = average phenotypic value of two parental populations.

Additionally, single-parent heterosis (genetic gain or improvement) is defined as the proportional increment in the phenotypic values of single-parent stock caused by crossing and calculated by the following equation (Cruz & Ibarra 1997):

[H.sub.x]% = FC - P(X)/P(X) x 100

where, [H.sub.x] = single-parent heterosis (genetic gain or improvement) of stock X; P(X) = phenotypic value of parental stock X; FC as defined before.

Statistical Analyses

Differences between parental stocks in fecundity, egg size (diameter) and shell length, as well as single-parent heterosis (genetic gain or improvement) for each stock were tested using the student t-test. Differences in hatching success, larval survival and growth among the four groups and in heterosis among different life history stages, ages, and types (midparent heterosis [H.sub.MP], single-parent heterosis for stock A [H.sub.A], and single-parent heterosis for stock B [H.sub.B]) were analyzed by multiple comparisons of means using a 1-way ANOVA. Shell length when appropriate was converted to logarithms to increase normality and homoscedasticity (Neter et al. 1985). Because data on hatching success, larval survival and heterosis were in percentages, they were transformed to arcsine before analysis (Rohlf & Sokal 1981). Analyses were done using SPSS (Statistical Program for Social Sciences) 11.5 software for Windows. Significance level for all analyses was set to P < 0.05 unless noted otherwise.

RESULTS

Differences Between the Two Parental Stocks

The two parental stocks were significantly different in body size and fecundity but not in egg size (Table 1 ). Body size of stock B as measured in shell height was 7% larger than that of stock A. Fecundity of stock B was 1.07 times higher than that of stock A.

Hatching

Hatching success of the hybrid cross AB was the highest, although it did not differ significantly from that of AA or the reciprocal hybrid cross BA (Fig. 1). The lowest hatching success was seen in the parental BB stock, but the differences among BB, AA and reciprocal hybrid cross BA was not significant. The difference between the best (AB) and worst (BB) performing groups was significant. Midparent heterosis for hatching success ([H.sub.MP] = 7.9%) fell between single-parent heterosis A ([H.sub.A] = 10.6%) and B ([H.sub.B] = 5.3%), all were significantly bigger than zero (P < 0.01). HA was significantly bigger than [H.sub.B] (P < 0.01) (Table 2).

[FIGURE 1 OMITTED]

Larval Survival

No significant differences (P > 0.05) for larval survival existed among four groups at days 4 and 7, but survival of hybrid AB were significantly higher than that of randomly mated BB (P < 0.05) at days 10 (Fig. 2). Midparent heterosis ([H.sub.MP]), single-parent heterosis A ([H.sub.A]) and single-parent heterosis B ([H.sub.B]) for larval survival were presented in Table 2. All heterosis was positive, and no significant differences were observed among them at the same age. Both [H.sub.MP] and [H.sub.A] increased significantly (P < 0.05) with increasing age, although HB showed no significant increase (P > 0.05). [H.sub.MP], [H.sub.A] and [H.sub.B] were all significantly bigger than zero (P < 0.01), which were 10.8%, 10.8% and 10.7%, respectively.

[FIGURE 2 OMITTED]

Growth

Shell length of the two parental stocks was significantly different from day 4 on, with the shell length of the BB stock always being larger than the AA stock (Table 3). Shell length of both bybrid crosses were intermediate between the two parental stocks and significantly larger than the shell length of the AA stock from day 7 to the end of the experiment (days 160). The BA cross was significantly larger than the AB cross and as large as the BB stock on days 4 and 30.

Mid-parent heterosis ([H.sub.MP]) for growth clearly fell between two single-parent heterosis and increased with age (Fig. 3). [H.sub.MP] at the larval stage (3.0%) was significantly smaller than that at spat (8.4%) (P < 0.05) and adult (10.1%) stages (P < 0.01) (Table 4).

[FIGURE 3 OMITTED]

For single-parent heterosis, positive values were observed for stock A throughout the studying period, whereas negative values were observed for stock B. Analysis of variance further demonstrated that the average of single-parent heterosis of stock A (23.0%) was significantly bigger than that of stock B (-4.0%) (P < 0.001, t = 9.221, df = 10). Moreover, single-parent heterosis of stock A varied significantly among the three life stages (Table 4), ranging from 9.8% at larval stage, to 29.8% at adult stages.

DISCUSSION

This study provides two major observations on heterosis in bay scallops. First, crosses between the two evaluated stocks of A. irradians irradians revealed midparent heterosis for all traits, although in no case where heterosis was useful (i.e., the hybrid crosses outperformed both parental stocks). Secondly, the magnitude of heterosis was not constant, but varied among traits and life history stages; and different traits exhibited different single-parent heterosis.

The observed differences in parental size and fecundity between stock A and B (Table 1) imply that there may be genetic differences between these stocks. Two reasons can potentially explain such results. First, the two parental stocks, A and B, have different origin and breeding history. The breeding history of stock A includes its reproduction under artificial conditions for 16 more generations than stock B. Because of this, it is expected that stock A has experienced a stronger accumulation of inbreeding, whereas stock B might be less inbred. Second, the origin of stock A and stock B are also different, with the former being derived from only 26 individuals (Zhang et al. 1986), whereas the later one was derived from 406 individuals, including 200 of wild origin and 206 from a one generation in captivity population (Gu, Z--one of the introducers, personal communication). Because of the small initial number of stock A and the mass spawning for so many generations, it is expected that stock A has accumulated high levels of inbreeding that is affecting fitness-related traits. The molecular evidence of reduced genetic diversity in stock A was provided by comparing with other natural populations (Blake et al. 1997). Stock B, however, may have retained high genetic diversity because of its shorter breeding history and larger initial population size.

Crosses between genetically differentiated subpopulations are expected to increase heterozygosity, reduce effects of recessive lethal genes and enhance fitness, resulting in heterosis or hybrid vigor (Whitlock et al. 2000). The hybrid crosses in this study exhibited positive midparent heterosis for hatching success, larval survival and growth. The presence of heterosis (positive or negative) has been reported for other marine bivalves (Mallet & Haley 1983, Hedgecock et al. 1995, 1996, Bayne et al. 1999) and other hermaphroditic pectinid species (Cruz & Ibarra 1997, Cruz et al. 1998, Beaumont et al. 2004). In the present study, the means for the hybrid crosses for all traits were always higher than that of the parental stocks, indicating a positive association between heterozygosity and the traits. Sheridan (1981) stated that the positive association is a common consequence of crossbreeding. Classical quantitative genetic studies of crossbreds produced by crossing inbred lines have uncovered remarkable heterosis in growth and its physiological components at larval, juvenile and adult stages and has implicated epistasis as a significant cause of this heterosis (Hedgecock et al. 1996).

In this study, we found the magnitude of heterosis is not constant but varies among traits and life stages. Midparental heterosis for fitness-related traits such as hatching success (7.9%) and larval survival (10.8%) is bigger than that for morphological traits such as larval shell length (3.0%). Similar results have previously been obtained by comparing pair-crosses between the two stocks with their self-fertilization in this species (Zheng et al. 2004b), and heterosis for larval survival (32.9%) was bigger than that for larval shell length (8.5%) at days 7. Why fitness-related traits such as survival exhibit greater heterosis than morphological traits? It is known that heterosis depends on the presence of directional dominance for the loci involved in the specific trait (Falconer 1981, Lamkey & Edwards 1998, 1999), and the coefficient of dominance variance is larger in fitness-related traits than in morphological traits (Roff 1998). Then, as stated by Lynch and Walsh (1998), fitness-related traits are expected to present directional dominance because mutations affecting those traits are typically deleterious and recessive. DeRose and Roff (1999) also emphasized that morphological traits exhibit little or no directional dominance. It is also known that life-stage specific difference in inbreeding depression are common and may depend on environmental, developmental or genetic factors (Husband & Schemske 1996). Therefore, complementary to the phenomenon of inbreeding depression, it is not uncommon for heterosis to be life-stage specific, as seen in this study, where midparent heterosis for growth was 3.0% at larval stage, 8.4% at juvenile stage and 10.1% at adult stage. It is also possible that heterosis is masked by maternal effects at early stages. Similar results have been reported in other marine bivalves. For example, in hard clam Mercenaria mercenaria, crossbred offspring were not consistently faster growing than purebred offspring (Manzi et al. 1991). In catarina scallop A. circularis, heterosis values for larval growth were 0 at day 11 and 6.8% at day 17 (Cruz & Ibarra 1997), and over 10% at adult stages (Cruz et al. 1998). In the Pacific oyster Crassostrea gigas, heterosis for body size were 1.0% at day 2 and 7.7% at day 340 (Hedgecock et al. 1995).

Finally, we found single-parent heterosis for growth between the two stocks was significantly different. The difference in single-parent heterosis (genetic gain or improvement) for growth between the two stocks could be caused by genetic differences between the two parental stocks. For stock A, performance may have been depressed by inbreeding and improved by crossing with stock B with higher genetic diversity; thus the value of single-parent heterosis (genetic gain or improvement) for growth is always positive. For stock B, there is little or no inbreeding depression, and the stock is performing better than stock A; thus the value of single-parent heterosis (genetic gain or improvement) for growth is negative.

ACKNOWLEDGMENTS

The authors thank the Scallop Seed Production Base of Hongchao Fishery Group, Center of Marine Shellfish Selection and Breeding of Laizhou City, for their assistance in hatchery and farm operations. This research was supported by grants from the Scientific Innovation Program of Chinese Academy of Sciences (ZKCX2-211 to Zhang) and National Science Foundation of China (30671622 to Zhang).

LITERATURE CITED

Bayne, B. L., D. Hedgecock, D. McGoldrick & R. Rees. 1999. Feeding behaviour and metabolic efficiency contribute to growth heterosis in

Pacific oyster. [Crassostrea gigas (Thunberg)]. J. Exp. Mar. Biol. Ecol. 233:115-130.

Beaumont, A. R., G. Turner, A. R. Wood & D. O. F. Skibinski. 2004. Hybridisations between Mytilus edulis and Mytilus galloprovincialis and performance of pure species and hybrid veliger larvae at different temperatures. J. Exp. Mar. Biol. Ecol. 32:177-188.

Blake, S. G., N. J. Blake, M. J. Oesterling & J. E. Graves. 1997. Genetic divergence and loss of diversity in two cultured populations of the bay scallop, Argopecten irradians (Lamarck, 1819). J. Shellfish Res. 16: 55-58.

Bruce, A. B. 1910. The Mendelian theory of heredity and augmentation of vigor. Science 32:627-628.

Crow, J. F. 1998. 90 year ago: the beginning of hybrid maize. Genetics 148:923-928.

Cruz, P. & A. M. Ibarra. 1997. Larval growth and survival of two catarina scallop (Argopecten circularis) populations and their reciprocal crosses. J. Exp. Mar. Biol. Ecol. 212:95-110.

Cruz, P., J. L. Ramirez, G. A. Garcia & A. M. Ibarra. 1998. Genetic differences between two populations of catarina scallop (Argopecten ventricosus) for adaptations for growth and survival in a stressful environment. Aquaculture 166:321-335.

DeRose, M. A. & D. A. Roff. 1999. A comparison of inbreeding depression in life-history and morphological traits in animals. Evolution Int. J. Org. Evolution 53:1288-1292.

East, E. M. 1936. Heterosis. Genetics 21:375-397.

Falconer, D. S. 1981. Introduction to Quantitative Genetics (3rd Edition). London and New York: Longman.

Griffing B. 1990. Use of a controlled-nutrient experiment to test heterosis hypotheses. Genetics 126:753-767.

Guo, X., S. E. Ford & F. S. Zhang. 1999. Molluscan aquaculture in China. J. Shellfish Res. 18:19-31.

Hedgecock, D., D. J. McGoldrick & B. L. Bayne. 1995. Hybrid vigor in Pacific oyster: an experimental approach using crosses among inbred lines. Aquaculture 137:285-298.

Hedgecock, D., D. J. McGoldrick, D. T. Manahan, J. Vavra, N. Appelmans & B. L. Bayne. 1996. Quantitative and molecular genetic analyses of heterosis in bivalve molluscs. J. Exp. Mar. Biol. Ecol. 203:49-59.

Husband, B. & D.W. Schemske. 1996. Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50:54-70.

Lamkey, K. R. & J. W. Edwards. 1998. Heterosis: theory and estimation. In: Proceedings 34th Illinois Corn Breeders' School, Urbana, 1L, 2-3 Mar. 1998. University of Illinois, Urbana, pp. 62-77.

Lamkey, K. R. & J. W. Edwards. 1999. The quantitative genetics of heterosis. In: J. G. Coors, & S. Pandey, editors. Proceedings of the International Symposium on the Genetics and Exploitation of Heterosis in Crops, CIMMYT, Mexico City, Mexico, Aug. 17-22, 1997. pp. 31-48.

Luo, L. J., Z. K. Li, H. W. Mei, Q. Y. Shu, R. Tabien, D. B. Zhong, C. S. Ying, J. W. Stansel, G. S. Khush & A. H. Paterson. 2001. Overdominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. II. Grain yield components. Genetics 158: 1755-1771.

Lynch, M. & B. Walsh. 1998. Genetics and analysis of quantitative traits. Sunderland, MA: Sinauer.

Mallet, A. L. & L. E. Haley. 1983. Growth rate and survival in pure population matings and crosses of the oyster Crassostrea virginica. Can. J. Fish. Aquacult. Sci. 40:948-954.

Manzi, J. J., N. H. Hadley & R. T. Dillon, Jr. 1991. Hard clam, Mercenaria mercenaria, broodstocks: growth of selected hatchery stocks and their reciprocal crosses. Aquaculture 94:17-26.

Neter, J., W. Wasserman & M. Kutner. 1985. Applied linear statistical models, 2nd. Ed. In: R. D. Iriwin, editor, pp 1127.

Roff, D. A. 1998. Effects of inbreeding on morphological and life history traits of the sand cricket, Gryllus firmus. Heredity 81:28-37.

Rohlf, F. J. & R. R. Sokal. 1981. Statistical tables. New York: W. H. Freeman and Company. pp 219.

Sastry, A. N. 1963. Reproduction of the bay scallop Aequipecten irradians Lamarck. Influence of temperature on maturation and spawning. Biol. Bull. 125:146-153.

Sheridan, A. K. 1981. Crossbreeding and heterosis. Anim. Breed. Abstr. 49:131-144.

Shull, G. H. 1908. The composition of a field of maize. Ann. Breed. Assn. 4:296-301.

Shull, G. H. 1914. Duplicate genes for capsule form in Bursa bursapastoris. Z. Indukt. Abstammungs Verebringsl. 12:97-149.

Shull, G. H. 1952. Beginnings of the heterosis concept.-48 In: J. W. Gowen, editor. Heterosis. Ames: Iowa State College Press. pp. 14.

Singh, S. M. & E. Zouros. 1978. Genetic variation associated with growth rate in the American oyster (Crassostrea virginica). Evolution Int. J. Org. Evolution 32:342-353.

Stiles, S. & J. Choromanski. 1995. Inbreeding studies on the bay scallop, Argopecten irradians. J. Shellfish Res. 14:278.

Stuber, C. W., W. P. Williams & R. H. Moll. 1973. Epistasis in maize (Zea mays L.): III. significance in predictions of hybrid performances. Crop Sci. 13:195-200.

Whitlock, M. C., P. K. Ingvarsson & T. Hatfield. 2000. Local drift load and the heterosis of interconnected populations. Heredity 84:452-457.

Wright, S. 1977. Evolution and the genetics of Population. Vol. 3: Experimental Results and Evolutionary Deductions. Chicago: The University of Chicago Press.

Zhang, F. S., Y. C. He, X. S. Liu, J. H. Ma & S. Y. Li. 1986. A report on the introduction, spat-rearing and experimental culture of bay scallop, Argopecten irradians Lamarck. Oceanologica et Limnologia Sinica 17:367-374. (in Chinese with English abstract).

Zhang, G. F., S. X. Liu, X. Liu, X. Guo & F. S. Zhang. 2003. Self-fertilization family establishment and its depression in bay scallop Argopecten irradians. J. Fish. Sci. China 10:441-445. (in Chinese with English abstract).

Zheng, H. P., G. F. Zhang, X. Liu, F. S. Zhang & X. Guo. 2004a. Different responses to selection in two stocks of the bay scallop, Argopecten irradians irradians Lamarck (1819). J. Exp. Mar. Biol. Ecol. 313:213-223.

Zheng, H. P., G. F. Zhang, X. Liu & H. Y. Que. 2004b. Comparison between the self-fertilized and hybridized families in Argopecten irradians irradians Lamarck, 1819. J. Fish. China 28:267-272. (in Chinese with English abstract).

HUAIPING ZHENG, (1) GUOFAN ZHANG, (1) * XIMING GUO (2) AND XIAO LIU (1)

(1) Key Laboratory of Experimental Marine Biology, Institute of Oceanology, Chinese Academy of Sciences, Qingdao, 266071, People's Republic of China; (2) Haskin Shellfish Research Laboratory, Institute of Marine and Coastal Sciences, Rutgers University, Port Norris, New Jersey 08349

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

TABLE 1.
Parental shell size, fecundity and egg size (diameter) of two stocks
of Argopecten irradians irradians. Standard deviation and samples
size are given in parenthesis.

              Parental                  Fecundity
Stock      Size (mm) (1)       (x [10.sup.4] eggs/spawner)

A       51.5 (a) (5.3, 200)       78 .0 (a) (17.9, 10)
B       54.9 (b) (10.4, 200)      161.l (b) (62.2, 10)

Stock   Egg Size ([micro]m)

A       58.0 (a) (2.8, 200)
B       58.3 (a) (3.2, 200)

(1) Within each column, means with the same letter are not
statistically different (P > 0.05).

TABLE 2.
Heterosis (%) for hatching success and larval survival in Argopecten
irradians irradians at days 4, 7 and 10. [H.sub.MP] is the mid-parent
heterosis. [H.sub.A] and [H.sub.B] are the single-parent heterosis of
stock A and B, respectively.

                                       Traits

                                  Larval Survival (1)
                   Hatching
Heterosis         Success (1)            Day 4

[H.sub.MP (%)    7.9 (ab) (3.6)      6.0 (a) (3.0)
[H.sub.A] (%)   10.6 (a) (3.3)       2.8 (a) (7.2)
[H.sub.B] (%)    5.3 (b) (4.8)       9.2 (a) (1.4)

                             Traits

                       Larval Survival (1)

Heterosis            Day 7            Day 10

[H.sub.MP (%)   11.2 (ab) (3.7)   15.1 (b) (1.9)
[H.sub.A] (%)   12.3 (ab) (3.2)   17.3 (b) (3.3)
[H.sub.B] (%)   10.2 (a) (4.9)    12.8 (a,b) (5.1)

(1) Means that share the same letter within a column (for hatching
success) or row (for larval survival) are not statistically different
(P > 0.05).

TABLE 3.
Shell length (SD) of the four genetic groups in Argopecten irradians
irradians at different ages. AA and BB are the two parental groups
(A[female] x B[male] and B[female] x A[male]).

                      Experimental Genetic Groups (1)
    Stage and
       Age                 AA                  AB
Larvae ([micro]m)
        1            90.1 (a) (4.3)      91.0 (a) (2.9)
        4           112.3 (a) (8.9)     116.3 (a) (7.9)
        7           131.0 (a) (12.3)    139.3 (b) (10.0)
       10           148.8 (a) (15.9)    168.5 (b) (17.4)
    Spat (mm)
       20           0.241 (a) (0.039)   0.287 (b) (0.053)
       30           0.637 (a) (0.142)   0.760 (b) (0.112)
       40           1.932 (a) (0.348)   2.392 (c) (0.346)
       50           3.923 (a) (0.794)   4.997 (b) (0.808)
   Adult (mm)
       70           11.12 (a) (2.78)    14.34 (b) (2.42)
       100          20.02 (a) (2.80)    25.44 (b) (2.95)
       130          29.78 (a) (5.33)    38.02 (b) (3.64)
       160          39.42 (a) (6.21)    48.49 (b) (5.79)

                       Experimental Genetic Groups (1)
    Stage and
       Age                 BA                  BB
Larvae ([micro]m)
        1            90.8 (a) (3.6)      91.5 (a) (4.0)
        4           122.3 (b) (7.5)     124.O (b) (11.4)
        7           145.l (bc) (11.7)   147.0 (c) (18.1)
       10           172.7 (b) (16.6)    174.3 (b) (22.14)
    Spat (mm)
       20           0.301 (b) (0.055)   0.312 (b) (0.072)
       30           0.854 (c) (0.092)   0.870 (c) (0.158)
       40           2.494 (c) (0.344)   2.524 (c) (0.483)
       50           5.134 (b) (0.718)   5.255 (b) (1.120)
   Adult (mm)
       70           14.70 (b) (2.16)    15.13 (b) (2.86)
       100          27.33 (bc) (2.92)   28.13 (c) (3.66)
       130          39.43 (b) (3.60)    40.35 (b) (6.05)
       160          51.46 (b) (4.89)    50.45 (b) (8.19)

(1) Within each row, means with the same letter are not statistically
different (P > 0.05).

TABLE 4.
Average heterosis (SD) for growth in Argopecten irradians irradians
during three different life stages. [H.sub.MP] is the mid-parent
heterosis. [H.sub.A] and [H.sub.B] are single-parent heterosis for
stock A and B, respectively.

                                   Stage

Heterosis            Larvae             Spat            Adult

[H.sub.MP] (%)    3.0 (ab) (2.8)    8.4 (c) (1.9)   10.9 (c) (1.1)
[H.sub.A] (%)     9.8 (a)  (4.3)   26.1 (d) (3.0)   29.8 (d) (2.1)
[H.sub.B] (%)    -3.0 (b)  (0.9)   -5.0 (b) (1.9)   -3.8 (b) (2.2)

(1) Within each row or line, means with the same letter are not
statistically different (P > 0.05).