Growth and survivorship of non-native (Crassostrea gigas and Crassostrea ariakensis) versus native eastern oysters (Crassostrea virginica).

ABSTRACT The decline of wild populations of the eastern oyster Crassostrea virginica from fishing impacts and disease combined with limited success in its culture has stimulated discussion among coastal managers about the risks and benefits of introducing non-native oysters in Maryland, Virginia, and North Carolina. Field experiments in 1999 to 2000 and 2001 to 2002 comparing growth, survivorship, and prevalence of disease in 2 non-native oysters, C. gigas and C. ariakensis, versus C. virginica in North Carolina estuaries demonstrated that in high-salinity (>25[per thousand]) waters, performance of C. gigas in culture greatly surpassed that of both of the other oysters (with growth 162.4% higher than C. virginica and 54.1% higher than C. ariakensis and survivorship 33.1% higher than C. virginica and 22.3% higher than C. ariakensis). C. ariakensis survivorship at these high salinity sites was highly variable and unpredictable even when using environmental covariates, and at salinities below ~10[per thousand] this species did not grow, rendering its culture nonviable at low salinity. However, in waters of intermediate salinity (15[per thousand] to 25[per thousand]), C. ariakensis outgrew both of the other 2 oysters (35.9% higher than C. gigas and 24.5% higher than C. virginica) and exhibited 42.1% higher survivorship than C. gigas. Although survivorship of C. virginica and C. ariakensis did not differ significantly at intermediate salinities, only C. virginica failed to achieve legally harvestable sizes and, based on its increasingly high susceptibility to death from disease with age, is likely to have experienced much greater mortality by the time of complete grow-out. Experimental elevation above the bottom augmented growth and survivorship of C. ariakensis most strongly, whereas C. gigas was not influenced by rack height. Before large-scale introduction of any non-native oyster occurs, the quantitative biologic results should first be incorporated into economic evaluations that weigh expected profitability and ecosystem benefits against the potential ecologic risks of introduction (both for wild release and for aquaculture of triploids).

KEY WORDS: Crassostrea ariakensis, Crassostrea gigas, Crassostrea virginica, economic feasibility, triploid, oyster disease, oyster growth, oyster survivorship

INTRODUCTION

Previous introductions of non-native species have often had severe consequences for ecologic communities, including reduction of the diversity, abundance and distribution of native fauna and flora (Carlton 1992, Ruiz et al. 2000). Consequently, prior to intentionally introducing any exotic fisheries species, managers should carefully weigh potential negative against positive ecological effects and evaluate under what conditions (aquaculture of nonreproductives versus wild release), if any, introduction may be economically and ecologically justifiable. A key component of this evaluation process is defining the rationale or need for introduction (Carriker 1992). Specifically, there must be clearly identified and scientifically defensible reasons why the native species is inadequate (Courtney & Robins 1989) and the introduced species is expected to have a high potential for success (Mann 1979, Rosenfield & Kern 1979, Mann et al. 1991).

Frustration with the slow pace of restoration efforts targeting wild populations of the eastern oyster Crassostrea virginica (Gmelin 1791) coupled with high mortality rates associated with culture of this species over the past several decades have resulted in advocacy by the shellfish industry to introduce non-native oysters in Maryland, Virginia, and North Carolina (Mann et al. 1991, Byrne 1996, Shatkin et al. 1997). Two species, the Pacific oyster, C. gigas (Thunberg 1793), and the Suminoe oyster, C. ariakensis (Fujita 1913), have been proposed as candidates for triploid aquaculture and even wild introductions. Native to Japan and the Korean peninsula (Mann et al. 1991), C. gigas has been successfully introduced to France, Oregon, Washington, western Canada, Australia, and New Zealand (Shatkin et al. 1997) and currently accounts for over 80% of the world's fishery production of oysters (Ayers 1991). Despite some taxonomic confusion with C. rivularis, the native distribution of C. ariakensis is believed to range from Pakistan through China to Japan, where it extends well into lower-salinity (i.e., <25[per thousand]) portions of estuaries (Breese & Malouf 1977, Langdon & Robinson 1996).

Resource managers currently face one of the most ecologically critical decisions in the history of environmental and fisheries management in the United States, whether C. gigas or C. ariakensis should be either intentionally released to propagate in the wild or cultured as nonreproductives in controlled aquaculture settings (National Research Council 2003). Some past studies provide information on and discussions of potential ecologic risks and perceived ecosystem (e.g., enhanced bio-filtration rates) and fisheries benefits of the 2 types of introduction (Mann 1979, Andrews 1980, Mann et al. 1991, Gaffney & Allen 1992, Lipton et al. 1992, Byrne 1996, Gottlieb & Schweighofer 1996, Shatkin et al. 1997). Several scientists have emphasized that significant risks to local and regional ecosystems exist and have yet to be fully addressed. For instance, introduction of reproductively viable non-native oysters could lead to eventual invasion of other estuaries in neighboring states or regions of the United States. Because information on the biology of these 2 species is sparse, the NRC Committee on Non-native Oysters in the Chesapeake Bay recently recommended that further research be conducted on the performance of native versus non-native oyster species (National Research Council 2003).

Realization of the potential fisheries benefits of introducing non-native oysters depends on their biology within the estuaries of the eastern United States, their marketability (see Grabowski et al. 2003 for relevant comparative information on marketability), and the integrated bioeconomics. Previous studies performed in Chesapeake Bay comparing the biology of C. virginica to either C. ariakensis or to C. gigas (Barber & Mann 1994, Calvo et al. 1999, Calvo et al. 2001) have documented higher resistance to Perkinsus marinus and Haplosporidium nelsoni and faster individual growth rates of both non-native species compared with C. virginica (Langdon & Robinson 1996, Calvo et al. 1999, Calvo et al. 2001), although the growth advantage seems to vary with salinity for C. gigas. However, direct comparison of the two non-native oyster species has yet to be conducted within the eastern United States and neither non-native species has been experimentally cultured in North Carolina. Here we present results of field trials covering the full range of potentially viable salinity regimes in coastal North Carolina. These trials were designed to assess the growth, survivorship, and disease prevalence and severity of the two non-native species under consideration for introduction and contrast these results to those obtained simultaneously for C. virginica. In addition, we tested if elevating oysters off the bottom differentially affects survival and growth of native versus non-native oysters.

MATERIALS AND METHODS

A series of experiments was conducted to compare growth, survivorship, and disease (dermo, P. marinus) prevalence and severity among native C. virginica (eastern oyster) and 2 non-native species, C. gigas (Pacific oyster) and C. ariakensis (Suminoe oyster), in North Carolina from 1999 to 2002. For each experiment, triploid seed C. gigas and C. ariakensis were obtained from the quarantine hatchery at the Virginia Institute of Marine Sciences (VIMS) and compared with diploid C. virginica obtained from Leslie Lee, Sloop Point Seafood, Hampstead, North Carolina. Triploid non-native oysters were raised at VIMS until achieving approximately 2-6 cm shell height (SH) in size, and tested for ploidy and disease status. Disease-free triploid oysters were shipped to the University of North Carolina-Institute of Marine Sciences (UNC-IMS), where they were held in upwellers prior to initiating field trials. Native oysters were raised at Sloop Point Seafood in raceways until the inception of each experimental trial. Oysters were then cultured in 4.8-mm mesh polyethylene bags that were elevated above the bottom using racks constructed from iron bars and located at approximately 0.1-0.5 m below mean low water (MLW) in research sanctuaries throughout coastal North Carolina.

First Series of Experiments (1999-2000)

Grow-out experiments were performed in 1999 to 2000 to compare growth and survivorship of C. virginica versus C. ariakensis and C. virginica versus C. gigas in separate experiments. In December 1999, C. ariakensis (mean SH [+ or -] 1 SE = 54.8 [+ or -] 0.9 mm) and C. virginica (42.6 [+ or -] 1.0 mm) were placed in 4.8-mm polyethylene (43 x 48 x 4 cm) bags on 15 cm-high racks at 1 high-salinity site (>25[per thousand]; Chadwick Bay) and 1 site with low (<10[per thousand]; Broad Creek) salinity (Fig. 1, Table 1). Abnormal environmental conditions following Hurricane Floyd resulted in extremely low salinity levels at Broad Creek in 2000 (Peterson 2000). Three bags of each species were deployed at each site with 52 oysters per bag. Living oysters were subsequently measured (SH) and counted to assess size and survivorship in March, June, and September/October 2000 at both sites.

[FIGURE 1 OMITTED]

C. gigas (31.4 [+ or -] 0.8 mm) and C. virginica (29.5 [+ or -] 0.6 mm) were placed in 6 polyethylene bags (50 oysters per bag, 3 bags per species) and deployed in February 2000 on 15 cm-high racks at each of 2 high-salinity sites in Waters Bay and Chadwick Bay (see Fig. 1). Deployment time for C. gigas differed from that of C. ariakensis described earlier because of availability of hatchery seed. C. gigas was not planted at Broad Creek because its poor performance at low salinities is already well documented (Calvo et al. 1999). Living oysters for the C. virginica/C. gigas contrast were subsequently measured (SH) and counted in May and August 2000 at both sites. For both the C. virginica/C. ariakensis and C. virginica/C. gigas experiments, oyster bags were washed with pressurized water and scrubbed with wire brushes to remove accumulated mud and fouling organisms during each sampling visit.

Second Series of Experiments (2001-2002)

The second series of experiments was initiated in April 2001 and included comparisons of growth, survivorship, P. marinus infection and Polydora spp. infestation among C. ariakensis, C. gigas, and C. virginica. Three bags of 70 C. ariakensis (31.8 [+ or -] 1.0 mm) and 3 bags of 70 C. virginica (20.6 [+ or -] 0.3 mm) were placed at each of 4 high-salinity (>25[per thousand]; Topsail Sound, Waters Bay, Chadwick Bay and Newport River, see Fig. 1) and 3 intermediate-salinity (15[per thousand] to 25[per thousand]; Bay River, Broad Creek and Swan Quarter) sites. In addition, 3 bags of 70 C. gigas (18.7 [+ or -] 0.3 mm) were placed at each of the high- and 1 intermediate-salinity (Bay River) sites to determine if slightly reduced salinities negatively impact C. gigas growth and survivorship in North Carolina. Bags (4.8-mm mesh, 43 x 48 x 4 cm) containing each set of seed oysters were placed on 15 cm-high rebar racks 0.1-0.5 m below MLW. To determine if height above the bottom affects native or non-native oyster growth and survivorship, 3 bags of C. virginica, 3 bags of C. gigas, and 3 bags of C. ariakensis were planted on racks at each of 2 additional heights (on the seabed and 38 cm above the seabed) at 2 sites (Chadwick Bay and Newport River).

Living oysters at high salinities were subsequently measured (SH) and counted in June, October, and November 2001. Living oysters at intermediate salinities were measured (SH) in June and October 2001, and February 2002. Salinity ([per thousand]), dissolved oxygen (mg/L), and water temperature ([degrees]C) at 0-25 cm beneath the water surface were measured monthly at each site from May to August 2001 during the experiment. Bags were cleaned with pressurized water and brushes monthly during the summer and seasonally during the fall and winter.

At each site, up to 24 oysters (4-8 oysters from each bag per species) were tested in August and October 2001 for prevalence and intensity of the oyster disease P. marinus and shell infestation rates by the mud worm Polydora spp. When testing for P. marinus infections, a 3-5-mm-long section of the rectum was removed from each oyster and analyzed for the presence and intensity of P. marinus using Ray's fluid Thioglycollate medium (RTFM) assays (Ray 1952, Ray 1963, Paynter & Burreson 1991). Infection intensity was calculated using the method described by Ray (1954) and Mackin (1962), with infection intensity categorized into the following groups: (0) absent, (1) light, (3) moderate, (5) heavy (Calvo et al. 1999, Lenihan et al. 1999). Average weighted intensity of P. marinus then was calculated for each species at each site by multiplying the number of oysters with each infection level by its infection intensity and dividing this sum by the total number of oysters tested. Intensity of Polydora spp. shell infestation was rated on a scale of 0 to 4 to describe the approximate percentage of the external oyster shell (right valve only) covered by mud worm tubes ([0] absent, [1] <25%, [2] 25% to 50%, [3] 50% to 75%, and [4] >75% coverage). Weighted intensity was calculated by multiplying the number of oysters with each intensity level by their respective infestation intensity and dividing the sum by the total number of oysters tested.

Statistical Analyses

A series of multifactor ANOVAs was performed to test if growth and survivorship varied between or among oyster species and whether these patterns were modified by grow-out site. Cochran's test for homogeneity of variances was conducted on all main effects prior to each analysis (Underwood 1981). For datasets that violated this assumption at [alpha] of 0.05, fourth-root transformations were performed and transformed data were tested. For experiments conducted in 1999 and 2000, separate 2-way ANOVAs were performed on oyster growth (final SH minus initial SH) and survivorship with site (Chadwick Bay and Broad Creek) and species (C. ariakensis and C. virginica) as fixed factors. A second set of separate 2-way ANOVAs was performed on oyster growth and survivorship with site (Chadwick Bay and Waters Bay) and species (C. gigas and C. virginica) as fixed factors. For experiments initiated in 2001, we conducted a series of analyses to compare growth and survivorship of the 3 oyster species within each salinity regimen. At relatively high salinities, we analyzed the effects of site (Chadwick Bay, Newport River, Topsail Sound, and Waters Bay) and species (C. ariakensis, C. gigas, and C. virginica) on growth and survivorship using separate 2-way ANOVAs with fixed factors. The effect of species (C. ariakensis, C. gigas, and C. virginica) on oyster growth and survivorship at Bay River (intermediate salinity) was analyzed using separate 1-way (fixed factor) ANOVAs. For the other 2 intermediate-salinity sites, 2-way ANOVAs were conducted on growth and survivorship with site (Swan Quarter Bay and Broad Creek) and species (C. ariakensis and C. virginica) as fixed factors. To test the effect of elevating oysters on their growth and survivorship, a 3-way ANOVA was performed with site (Chadwick Bay and Newport River), species (C. ariakensis, C. gigas, and C. virginica), and elevation (bottom, low, and high) as fixed factors. Student-Newman-Keuls (SNK) post hoc tests at [alpha] of 0.05 were conducted on all main effects. If an interaction proved significant in 2-way ANOVAs, SNK tests were performed among treatments within each level of a factor. The SNK test was selected because we conducted a balanced experiment with a priori predictions and fixed factors (Day & Quinn 1989).

RESULTS

First Series of Experiments

Results of the contrasts between C. ariakensis and C. virginica varied between the (low-salinity) Broad Creek and (high-salinity) Chadwick's Bay sites. C. ariakensis deployed in December grew by September from 55.3 mm SH to 56.9 [+ or -] 0.6 mm (mean [+ or -] 1 standard error) at Broad Creek and from 54.3 to 99.3 [+ or -] 1.9 mm at Chadwick Bay, while C. virginica increased during this period from 42.5 to 51.0 [+ or -] 2.2 mm at Broad Creek and from 42.6 to 71.0 [+ or -] 0.9 mm at Chadwick Bay. ANOVA revealed a significant effect of the interaction between site and species ([F.sub.1.8] = 79.1, P < 0.0001; Fig. 2a) on oyster growth (i.e., change in shell height). C. ariakensis grew 16.6 mm more than C. virginica at the high-salinity Chadwick Bay, but C. virginica outgrew C. ariakensis by 6.8 mm at the low-salinity Broad Creek (SNK post hoc comparisons; see Fig. 2a). ANOVA revealed that the interaction between site and species did not affect oyster survivorship ([F.sub.1,8] = 0.3, P = 0.58). Oyster survivorship at Chadwick Bay was 16.5% higher than at Broad Creek (site effect: [F.sub.1,8] = 28.4, P = 0.0007), and survivorship of C. virginica was 9.4% higher than that of C. ariakensis (species effect: [F.sub.1,8] = 9.1, P = 0.02; see Fig. 2b).

[FIGURE 2 OMITTED]

Results of the contrasts between C. gigas and C. virginica did not vary between sites, both of which had similarly high salinities. By the end of August 2000, C. gigas deployed in January 2000 grew from 31.3 to 81.3 [+ or -] 1.1 mm SH at Chadwick Bay and from 31.5 to 90.2 [+ or -] 2.1 mm at Waters Bay, whereas C. virginica grew from 29.8 to 48.8 [+ or -] 1.7 mm at Chadwick Bay and from 29.1 to 54.8 [+ or -] 1.7 mm at Waters Bay. For the contrast between C. gigas and C. virginica, ANOVA revealed no significant effect of the interaction between site and species on either growth ([F.sub.1,8] = 0.4, P = 0.56) or survivorship ([F.sub.1,8] = 0.4, P = 0.53). C. gigas grew more than C. virginica at both sites (species effect: [F.sub.1,8] = 455.3, P < 0.0001), and oyster growth was higher at Waters Bay for both species (site effect: [F.sub.1,8] = 26.6, P = 0.0009; see Fig. 2c). Survivorship of C. gigas was 18.7% higher than that of C. virginica across both sites (species effect: F1.8 = 8.8, P = 0.02; see Fig. 2d). Thus, in these high-salinity sites C. gigas grew faster and survived better than the native oyster.

Second Series of Experiments

Salinity, Temperature, and Dissolved Oxygen

Physical parameters were quantified in the summer of 2001 to indicate how variation in these factors might influence patterns of oyster growth and mortality. Between May and August 2001, mean salinity was 38[per thousand] at Chadwick Bay, Waters Bay, and Topsail Sound and 34[per thousand] at Newport River sites (Table 1). Mean dissolved oxygen ranged from 5.4 to 6.6 mg/L and mean water temperature from 24.0[degrees]C to 26.8[degrees]C during this period at the high-salinity sites (Table 1). Between May and August 2001, mean salinity was 18%o, dissolved oxygen 5.5 mg/L, and water temperature 26.7[degrees]C at Bay River (Table 1). Mean salinity at Swan Quarter Bay (16[per thousand]) was slightly lower than at Broad Creek (18[per thousand]) between May and August 2001 (Table 1). Mean dissolved oxygen was higher than all other sites at Swan Quarter Bay (7.2 mg/L), and this was the only site for the entire study where relatively low (<4.5 mg/L) dissolved oxygen levels were never recorded (Table 1). Finally, mean water temperature was slightly higher at Swan Quarter Bay (28.6[degrees]C) than at Broad Creek (27.1[degrees]C; Table 1).

High-salinity Sites

C. ariakensis versus C. gigas versus C. virginica. Growth and survivorship differed among non-native and native oysters in this set of trials. From April to November 2001, C. gigas in high salinity grew from 19.2 to 101.4 [+ or -] 2.3 mm SH (means of all 4 sites), C. ariakensis from 31.6 to 86.0 [+ or -] 2.2 mm, and C. virginica from 20.7 to 52.4 [+ or -] 1.9 mm. The interaction between site and species did not affect oyster growth ([F.sub.6,24] = 2.0, P = 0.11), but each main effect was significant (site: [F.sub.3,6] = 25.8, P < 0.0001; species: [F.sub.2,6] = 346.3, P < 0.0001). Oyster growth at Waters Bay was greater than all other sites, which did not differ (SNK post hoc comparisons; Fig. 3a). C. gigas growth was greater than that of the other 2 species, and C. ariakensis outgrew C. virginica (see Fig. 3b). ANOVA revealed a significant interaction of site and species for oyster survivorship ([F.sub.6,24] = 12.8, P < 0.0001; see Fig. 3c). Survivorship of C. gigas exceeded that of C. ariakensis at the Chadwick Bay and Newport River sites, and was greater than that of C. virginica at all 4 sites (Fig. 3c). Survivorship of C. ariakensis exceeded that of C. virginica at Topsail Sound and Waters Bay, but was less than that of C. virginica at Newport River (Fig, 3c). Finally, survivorship of C. ariakensis and C. virginica did not differ at Chadwick Bay (Fig. 3c).

[FIGURE 3 OMITTED]

Intermediate-salinity Sites

Bay River (C. ariakensis versus C. gigas versus C. virginica). In contrast to the results at the high-salinity sites, both C. ariakensis and C. virginica exhibited higher growth and survivorship than C. gigas at Bay River. From April 2001 to February 2002, C. ariakensis had grown from 32.5 to 82.9 [+ or -] 1.7 mm SH, C. virginica from 20.1 to 56.3 [+ or -] 0.02 mm, and C. gigas from 16.5 to 43.2 [+ or -] 1.1 mm. C. ariakensis outgrew both of the other species, and growth of C. virginica was greater than that of C. gigas (SNK post hoc comparisons; [F.sub.2,6] = 88.0, P < 0.0001; Fig. 4a). Survivorship did not differ between C. ariakensis and C. virginica, but survivorship of each was greater than that of C. gigas ([F.sub.2,6] = 52.4, P = 0.0002; see Fig. 4b).

[FIGURE 4 OMITTED]

Broad Creek and Swan Quarter Bay (C. ariakensis versus C. virginica). The pattern exhibited at the other intermediate-salinity site (Bay River) of higher growth of C. ariakensis than C. virginica but equivalent survivorship was replicated in this set of trials. From April 2001 to February 2002, C. ariakensis grew from 31.4 to 62.1 [+ or -] 1.0 mm SH at Broad Creek and from 31.1 to 80.7 [+ or -] 3.5 mm at Swan Quarter Bay, whereas C. virginica increased from 18.8 to 48.8 [+ or -] 1.0 at Broad Creek and from 18.9 to 56.5 [+ or -] 0.6 mm, respectively, at the 2 sites. ANOVA revealed a significant effect of the interaction between site and species for oyster growth ([F.sub.1,8] = 7.5, P = 0.03; Fig. 4c). C. ariakensis outgrew C. virginica at Swan Quarter Bay, but not at Broad Creek (SNK post hoc comparisons; Fig. 4c). For both species, growth was greater at Swan Quarter Bay than at Broad Creek (Fig. 4c). ANOVA revealed no significant effect on oyster survivorship of either main effect (site: [F.sub.1,8] = 0.5, P = 0.51; species: [F.sub.1,8] = 1.4, P = 0.27) or the interaction between site and species ([F.sub.1,8] = 2.8, P = 0.13). Mean survivorship at these 2 sites was 74.1 [+ or -] 4.2 for C. ariakensis and 83.4 [+ or -] 7.3 for C. virginica.

Oyster Elevation Experiment

C. ariakensis versus C. gigas versus C. virginica. Elevating oysters enhanced the growth of C. ariakensis more than that of the other two oyster species. ANOVA revealed a significant effect of the 3-way interaction among site, species, and elevation for oyster growth (Table 2 and Fig. 5a). Height of oysters did not affect C. gigas growth at either site (SNK post hoc comparisons; Table 2, Table 3 and Fig. 5a). C. ariakensis growth was greater on high racks than on the bottom at both sites, and was also greater on high-rack racks than on low racks at Chadwick Bay (Table 2, Table 3, and Fig. 5a). C. virginica growth was greatest on high racks, intermediate on low racks, and lowest on the bottom at Newport River, but did not differ at Chadwick Bay (Table 2, see Fig. 5a). C. gigas growth was greater than that of the other 2 species for all 3 elevations at both sites (Table 2, Table 3, and Fig. 5a). C. ariakensis growth was greater than that of C. virginica at all 3 elevations at both sites except for the bottom at Chadwick Bay, where the 2 species did not differ (see Table 2 and Fig. 5a).

[FIGURE 5 OMITTED]

Although ANOVA of oyster survivorship revealed no 3-way interaction among site, species and elevation, all 3 2-way interactions were significant (site X species, species X elevation) or marginally significant (site X elevation; Table 2 and Fig. 5b,c). For the interaction between site and species, C. gigas survivorship was significantly higher than that of the other 2 species at both sites (SNK post hoc comparisons; Table 2 and Fig. 5b). C. ariakensis survivorship was greater than that of C. virginica at Chadwick Bay, but survivorship of the 2 species did not differ at Newport River (Table 2 and Fig. 5b). For the interaction between species and elevation, elevating oysters did not affect survivorship of C. gigas (Table 2 and Fig. 5c). In contrast, elevating C. ariakensis from the bottom or low racks to high racks increased its survivorship (Table 2 and Fig. 5c). Survivorship of C. virginica was greater on high racks than on the bottom, but did not differ from low racks (Table 2 and Fig. 5c). C. gigas survivorship was greater than that of the other 2 species at all 3 elevations (Table 2 and Fig. 5c). Survivorship was greater for C. ariakensis than for C. virginica only when oysters were raised on high racks (Table 2 and Fig. 5c). For the marginally significant (Table 2, P = 0.05) interaction between site and elevation, elevating oysters affected oyster survivorship only at Newport River, where survivorship of oysters was significantly greater on high racks than on the bottom.

Incidence of Oyster Disease and Polydora spp. Infestation

The prevalence (% of oysters infected) and intensity of P. marinus infection was extremely low at all sites for oysters tested in both August (1.7% were infected) and October 2001 (3.3 % were infected; Table 4). In August, P. marinus was detected at only 1 to 2 sites for each of the 3 species, and its average prevalence for any species at any site was never greater than 10.0% (Table 4). In October, P. marinus was detected in C. ariakensis at 4 of 7 sites, and was most prevalent (16.7%) among C. ariakensis on low racks in Newport River and most intense (0.67) among C. ariakensis at Topsail Sound (Table 4). Of the 3 species, P. marinus was least prevalent and least intense among C. gigas and was detected only among C. gigas on high racks at Chadwick Bay in October (Table 4). P. marinus was detected in C. virginica at 4 of 7 sites in October (see Table 4).

In August 2001, the prevalence (% of oysters infected) and intensity of Polydora spp. infestation were greatest among C. ariakensis, intermediate among C. gigas, and almost nonexistent for C. virginica (Table 5). Prevalence and intensity of mud worms on C. ariakensis shells were very high at Broad Creek and Swan Quarter Bay, intermediate at Chadwick Bay, and very low at the other 4 sites (Table 5). Mud worm tubes were present on over half of C. gigas oysters tested at Chadwick Bay, but were present at only 1 of the other 4 sites (6.7% at Waters Bay; see Table 5). Waters Bay was the only site where mud worm tubes were present on C. virginica oysters (Table 5). In October 2001, mud worm prevalence and intensity were greatest among C. ariakensis and C. gigas, and almost nonexistent among C. virginica. Mud worms were present on C. ariakensis and C. gigas shells at all sites except Waters Bay and Topsail Sound, and were most prevalent and intense at Chadwick Bay (Table 5). For C. virginica, mud worms were detected only at Chadwick Bay on high racks. In October, both infestation prevalence and intensity on shells of C. ariakensis and C. gigas were slightly greater on low racks than on either the bottom or high racks.

DISCUSSION

Critical to any decision on the introduction of non-native species for aquaculture, fisheries, or restoration of ecosystem services once provided by native species is an assessment of the biology of the candidate species in their prospective new environment. The primary motivations for introducing 1 of the 2 non-native oysters to Chesapeake Bay are their presumed resistance to P. marinus and H. nelsoni, with consequent survival advantages over the native oyster, and their high individual growth rates (National Research Council 2003). Previous studies comparing non-native oysters to C. virginica have found that the 2 non-native species tend to grow and survive better than the native oyster (Barber & Mann 1994, Calvo et al. 1999, Calvo et al. 2001), although reducing salinity can decrease or eliminate the growth and survivorship advantages of C. gigas. Barber and Mann (1994) demonstrated that growth and survivorship of C. gigas grown in the Chesapeake Bay were negatively impacted by salinities below 20[per thousand]. Similarly, Calvo et al. (1999) found that growth and survivorship of C. gigas in Chesapeake Bay were reduced at an intermediate salinity range of 15[per thousand] to 25[per thousand] in contrast to sites with salinities consistently above 25[per thousand]. They also noted that the individual growth rate of C. gigas was no longer greater than that of C. virginica at this intermediate-salinity level. Calvo et al. (2001) found low mortality rates for C. ariakensis over a wide spectrum of salinities (<15[per thousand], 15[per thousand] to 25[per thousand], >25[per thousand]), suggesting that the physiology of C. ariakensis is influenced less by salinity than that of C. gigas. High C. virginica mortality rates in 2 of these studies (Barber & Mann 1994, Calvo et al. 2001) were primarily caused by the parasitic protozoan P. marinus, which is one of the largest impediments to native oyster aquaculture and recovery of native oyster fisheries in the estuaries of Maryland and Virginia. Prior to our study, growth and survivorship of the two non-native species had yet to be compared along the Atlantic coast of the United States, though Robinson and Langdon (1993) found that C. gigas growth was greater than that of C. ariakensis at sites on the West coast.

Results of our study provide clear evidence that the 2 non-native oysters, C. gigas and C. ariakensis, differ dramatically from one another and from the native eastern oyster, C. virginica in critical biologic rates (Table 6). As anticipated from previous studies done in the Chesapeake Bay (Barber & Mann 1994, Calvo et al. 1999), we found that C. gigas grows faster and survives at higher rates in high-salinity waters (25[per thousand] to 36 [per thousand]) than in 15[per thousand] to 25 [per thousand] salinities. At the high-salinity sites, C. gigas exhibited consistent and substantial growth and survival advantages over the other 2 species (i.e., 162.4% higher growth than C. virginica and 54.1% higher than C. ariakensis and 33.1% higher survivorship than C. virginica and 22.3% higher than C. ariakensis). Thus, at high salinity, performance of the non-native C. gigas greatly exceeds that of the native eastern oyster in both biologic traits critical to production, namely growth and survivorship. From previous research (Barber & Mann 1994) and our own more limited data, the survivorship advantage of C. gigas could be related to greater resistance to P. marinus infection. At the 1 site of intermediate salinity (15[per thousand] to 25 [per thousand]) where we deployed C. gigas, it was significantly outperformed in both growth and survival by the native eastern oyster, C. virginica. Therefore, any enthusiasm for introduction of C. gigas to Chesapeake Bay or the Pamlico Sound must be tempered by the realization that in the vast majority of the waters of these estuaries salinities favor the native eastern oyster.

Our results from deploying C. ariakensis in the small-scale grow-out trials confirm some previous conclusions from analogous research in Chesapeake Bay (Calvo et al. 2001) while providing new insights as well from direct contrasts with C. gigas and from our tests of elevation impacts. We first demonstrated in cold-season trials that salinity levels below 10%o virtually inhibited all net growth and caused high mortality of C. ariakensis, thereby serving to help define one environmental and thus geographic limit to its successful culture. A site with salinity consistently below 10[per thousand] proved unsuitable to achieve net growth in winter and simultaneously induced high mortality. The native eastern oyster actually significantly outperformed C. ariakensis at this low-salinity site in both growth and survival. However, the native oyster would have required an additional 3 to 4 y to achieve market size based on observed growth rates of C. virginica from this site, rendering such environments poor candidates for its culture also.

At sites of intermediate salinity (15[per thousand] to 25 [per thousand]), C. ariakensis significantly outgrew C. gigas by 35.9% and C. virginica by 24.5%. At intermediate salinity, C. ariakensis survivorship was 42.1% higher than that of C. gigas, but it did not differ significantly from C. virginica. However, the absence of a survival advantage of C. ariakensis over C. virginica in our study is misleading. By the time our trials were terminated, C. ariakensis had already reached a legally harvestable size (76.2 mm SH in North Carolina), whereas the more slowly growing native oyster had not. Using observed growth rates of C. virginica in our study, approximately 2-10 additional months of culture would have been required to achieve market size. The several additional months of warm water exposure required to complete grow-out of C. virginica would almost certainly have elevated its mortality, perhaps even dramatically if P. marinus infection had increased as expected (Lenihan et al. 1999).

At sites of high salinity, growth of C. ariakensis consistently and significantly exceeded that of the native C. virginica in all 5 trials where this contrast was set up. However, C. ariakensis survivorship was highly variable and unpredictable even using the environmental information on actual salinity, temperature, and DO variation that we collected. Over the 5 trials comparing C. ariakensis to C. virginica at high salinity, C. virginica survived at a significantly higher rate in 2 cases, C. ariakensis survived better in 2 cases, and no significant difference was detected in the remaining contrast. The high variability in the survivorship results for these 2 species at high salinity differs from the consistent advantage of C. ariakensis previously demonstrated in the Chesapeake Bay study of Calvo et al. (2001).

Comparison of triploid non-native oysters with diploid C. virginica could partly explain why non-native oysters outgrew C. virginica because reduced gamete production in triploids generally results in enhanced somatic growth (Barber & Mann 1994). Allen and Downing (1986) and Davis (1989) documented that triploid C. gigas outgrow diploid C. gigas, particularly during the reproductive season. However, in a previous study with triploid C. gigas and C. virginica, C. gigas growth was nearly double that of triploid native oysters at high salinities (Calvo et al. 1999), suggesting that growth results in our study are only slightly confounded by differences in ploidy status among species.

A second potential limitation of this study was that the size and condition of oysters differed among species at the beginning of each experiment. In particular, C. ariakensis were approximately 10 mm (SH) larger than either of the other two species at the beginning of the experiment. C. ariakensis were raised at VIMS until they were large enough to be tested for ploidy status prior to use in this study. Because the proportional growth in oyster biomass increases with each incremental gain in shell height, an incremental gain in shell height for a larger oyster represents greater growth in biomass than the amount of biomass growth from a similar gain in shell height of a smaller oyster. Therefore, in this study growth rates of C. ariakensis are likely underestimated relative to the other two species. In particular, differences in growth rates between C. gigas and C. ariakensis were likely overestimated at high salinity sites, and C. ariakensis growth advantages over C. virginica and C. gigas (at low salinities) were probably underestimated. Another important consideration is that these oyster species differ in morphology, so that an incremental change in SH for each species does not necessarily represent a uniform change in oyster biomass. Comparison of oysters tissue weights of larger (80-110 mm SH) oysters that did not differ in SH determined that C. ariakensis tissue weight was approximately twice that of C. virginica and one-third greater than C. gigas tissue weight (Grabowski et al. 2003). Thus, comparing growth rates by quantifying changes in shell height also underestimated C. ariakensis growth relative to the other two species.

Our experiment that varied the elevation of oysters from culture on the bottom to racks of 2 different heights, 15 cm and 38 cm, provides some insight into why the relative advantage of C. ariakensis may change among sites even with salinity held constantly high. Varying elevation off the bottom bad no detectable impact on growth of C. gigas and did not change the survivorship advantage that C. gigas held over both other oysters. However, culturing C. ariakensis on the bottom consistently reduced its growth rate. Growth of C. virginica also exhibited lower growth in bottom culture at 1 of the 2 sites, but the native oyster was sufficiently less sensitive to the bottom environment such that the statistically significant growth advantage held by C. ariakensis over C. virginica at both rack elevations disappeared on the bottom at one site. Similarly, C. ariakensis held a detectable survivorship advantage over C. virginica only on high racks. Because the concentration of suspended sediments decreases dramatically with elevation in the water column in estuaries and suspended sediments can interfere with suspension feeding (Rhoads & Young 1970), our results from the manipulation of elevation of culture imply that C. ariakensis is more sensitive to elevated turbidity than the other two oysters. Consequently, variation in suspended sediment load may help explain the high variation in C. ariakensis survivorship among high-salinity sites. Such sensitivity to bottom culture implies that C. ariakensis may experience difficulty in becoming established in more turbid regions of estuaries.

Although results of our manipulation of culture elevation imply greater sensitivity of C. ariakensis to turbidity, it is doubtful that this explanation accounts for all the variation in its survivorship among high-salinity sites. Other factors varied among sites, such as extensive shell fouling by barnacles and tunicates at Newport River and Chadwick Bay, which may have contributed to mortality. Higher mortality rates could also be a consequence of the parasite Bonamia sp., which has caused extensive mortality among juvenile oysters in laboratory and field trials conducted at UNCIMS (Bishop et al. unpublished data). Rearing organisms in hatcheries often results in extreme genetic bottlenecks (Gaffney et al. 1996, Launey & Hedgecock 2001), which could increase cultured species' susceptibility to parasites and diseases. In the absence of the ability to predict mortality from known independent environmental variables that could be measured a priori at any prospective aquaculture site, the possibility of high mortality renders culture of C. ariakensis in high salinity a very risky proposition.

Creating any structures rising more than 15 cm off the bottom in North Carolina waters requires growers to obtain a water column lease in addition to the standard bottom lease. Because water column leases historically have been very difficult to obtain in North Carolina and cost an additional $100 per acre, advantages of using high racks must outweigh the added expense. Elevating oyster racks from 15 to 38 cm increased oyster growth and survivorship of C. ariakensis at both sites, growth of C. virginica at Newport River, and survivorship of C. virginica at both sites. The magnitude of the effects of increasing rack height from on the bottom and at 15 cm to 38 cm was greatest for C. ariakensis, whereas elevating oysters to 38 cm did not affect growth or survivorship of C. gigas at either site. Elevating oysters from the bottom to 15 cm generally did not affect growth or survivorship for any of the 3 species. Our results suggest that oyster growers culturing C. virginica or especially C. ariakensis, but not C. gigas, might consider obtaining a water column lease, though a complete bioeconomic evaluation of whether increased growth and survivorship outweigh the additional costs should be considered first.

Our results from trial culture comparing the performance of 2 non-native oysters to the native eastern oyster provide reasonably clear conclusions. C. gigas consistently outperforms both other oysters in growth and survivorship in high-salinity waters, but does less well than the other 2 oysters in intermediate salinity. At high salinity, C. ariakensis can suffer extremely high mortality, perhaps in part from exposure to high turbidity or the parasite Bonamia sp., but the environmental determinants are not well enough known to predict where survival will be good or bad. Consequently, on the basis of unpredictable and occasionally massive mortality, culturing C. ariakensis at sites of high salinity is risky. On the other hand, C. ariakensis has a substantial growth advantage over both of the other 2 oysters at intermediate salinity and likely also has a survivorship advantage. The range of viable salinities for successful culture of C. ariakensis does not extend below approximately 10[per thousand]; however, in the range of 15[per thousand] to 25[per thousand], this oyster grows faster and suffers less from the oyster diseases that plague C. virginica. Elevation of C. ariakensis during culture should be a viable strategy to increase survivorship and growth, but the cost-benefit ratio of obtaining water column leases should be examined further in North Carolina. Incorporation of all these results into a bioeconomic model is now necessary to quantify and compare the additional value that could be generated from higher growth and/or survivorship rates of non-native species relative to the risks associated with introducing a non-native species.

Because oyster consumers in eastern North Carolina prefer C. virginica over either non-native species when consumed raw and C. gigas when eaten cooked (Grabowski et al. 2003), differences (if any) in the price of each species for both raw and steamer markets must he incorporated into economic evaluation of the profitability of culturing non-native oysters. Given that the added cost of producing triploid oysters to avoid wild introduction is high and non-native oysters may be somewhat less palatable, culture of triploid non-native oysters may prove economically non-viable. Making non-native oysters available for aquaculture may, however, lower the probability of unsanctioned and uncontrolled introduction of reproductively capable non-native oysters into the environment (National Research Council 2003), thereby reducing the risk of potentially dire ecologic impacts of introducing a nonnative species (i.e., competition with native species, unintentional introduction of additional predators and/or diseases, etc.). Finally, potential ecosystem benefits (i.e., water filtration, habitat provision) of promoting bivalve aquaculture should also be considered in deciding about permits for culturing non-native oysters. The question of whether to attempt to establish breeding, self-replicating populations of a non-native oyster entails consideration of many more issues, but requires more biologic information on potential risks versus economic and ecosystem benefits.

TABLE 1.

Mean and range of salinity, dissolved oxygen, and temperature from 25
cm below the water surface at field (~0.1-0.5 m below MLW) sites in
North Carolina during the second experiment. Each parameter was
recorded monthly between May and August of 2001.

        Site             County (NC)         Latitude

Broad Creek, Wanchese     Dare          N35[degrees]50.530'
Swan Quarter Bay          Hyde          N35[degrees]23.217'
Bay River                 Pamlico       N35[degrees]11.032'
Newport River             Carteret      N34[degrees]44.689'
Chadwick Bay              Onslow        N34[degrees]31.603'
Waters Bay                Onslow        N34[degrees]26.941'
Topsail Sound             Pender        N34[degrees]24.417'

                                                  Salinity
                                                   ([per
                                                 thousand])

        Site                  Longitude         Mean    Range

Broad Creek, Wanchese    W75[degrees]37.170'     18     15-22
Swan Quarter Bay         W76[degrees]19.618'     16     15-16
Bay River                W76[degrees]36.529'     18     16-20
Newport River            W76[degrees]39.679'     34     31-36
Chadwick Bay             W77[degrees]22.574'     38     35-40
Waters Bay               W77[degrees]32.248'     38     35-40
Topsail Sound            W77[degrees]35.848'     38     35-40

                            Dissolved         Temperature
                          Oxygen (mg/L)       ([degrees]C)

        Site             Mean     Range     Mean      Range

Broad Creek, Wanchese    5.9     4.4-7.6    27.1    21.7-30.1
Swan Quarter Bay         7.2     6.3-8.3    28.6    24.5-30.5
Bay River                5.5     4.0-7.6    26.7    20.9-29.7
Newport River            5.4     4.0-6.3    25.9    21.0-28.2
Chadwick Bay             5.5     4.3-8.3    25.6    19.5-29.2
Waters Bay               6.6     4.4-9.5    24.0    19.6-27.8
Topsail Sound            5.8     3.7-7.7    26.8    19.3-30.8

TABLE 2.

The effect of site (Newport River and Chadwick Bay), species
(C. ariakensis, C. gigas, and C. virginica), and elevation (bottom,
low rack, and high rack) on oyster growth (change in shell height: SH)
and survivorship in 2001 analyzed using separate 3-way ANOVAs.

                                Oyster Growth (SH)

                           df     SS        F        P

Site                       1     0.001      3.6    0.06
Species                    2     0.302    522.4    <.0001
Height                     2     0.009     15.9    <.0001
Site x species             2     0.002      3.8    0.03
Site x height              2     0.001      2.0    0.16
Species x height           4     0.010      8.5    <.0001
Site x species x height    4     0.005      4.6    0.004
Residual                   36    0.010

                                Oyster Survivorship

                           df     SS        F        P

Site                        1    0.079      7.4    0.01
Species                     2    1.671     78.2    <.0001
Height                      2    0.340     15.9    <.0001
Site x species              2    0.333     15.6    <.0001
Site x height               2    0.069      3.2    0.05
Species x height            4    0.132      3.1    0.03
Site x species x height     4    0.019      0.4    0.78
Residual                   36    0.385

3-way interaction for oyster growth: site x species x height

C. ariakensis     D                 C               B
    C. gigas      A                 A               A
C. virginica      I                 H               F

                Bottom             Low            High

                             Newport River

                A > B    B = C    C = D   D = E   E = F
                A > C    B > D    C = E   D > F   E = G
                A > D    B > E    C > F   D > G   E > H
                A > E    B > F    C > G   D > H   E > I
                A > F    B > G    C > H   D > I
                A > G    B > H    C > I
                A > H    B > I
                A > I

C. ariakensis              D                D       B
    C. gigas               A                A       A
C. virginica               E                G       F
                         Bottom            Low    High

                               Chadwick Bay

                F = G    G = H    H > I
                F > H    G > I
                F > I

2-way interaction for oyster survivorship: site x species

Newport River         C                    A                    C
Chadwick Bay          B                    A                    C

                C. ariakensis           C. gigas           C. virginica

                   A > B        B = C
                   A > C

2-way interaction for oyster survivorship: species x height

C. ariakensis         E                    E                    B
    C. gigas          A                    A                    A
C. virginica          E                    D                    C

                   Bottom                 Low                  High

                    A > B       B > C    C = D     D = E
                    A > C       B > D    C > E
                    A > D       B > E
                    A > E

Notes: Provided also are results of SNK post hoc tests (using
Bonferroni's adjustment for multiple contrasts to maintain
experiment-wise [alpha] = 0.05) for each significant interaction at
P < 0.05 for the above analyses.

TABLE 3.

Initial and final oyster sizes (shell height) of oysters grown at each
elevation (bottom, low rack, and high rack) from April to November of
2001 at high-salinity sites (Newport River and Chadwick Bay) in North
Carolina.

                                    C. ariakensis           C. gigas

    Site          Elevation    Initial SH    Final SH      Initial SH

Newport River       Bottom        30.4       76.8 (0.6)       20.9
                    Low           28.9       83.9 (3.8)       19.6
                    High          30.4       89.9 (1.1)       20.9
Chadwick Bay        Bottom        30.9       71.5 (l.9)       18.5
                    Low           31.0       74.9 (2.8)       19.5
                    High          30.9       89.4 (1.1)       18.5

                                C. gigas             C. virginica

    Site          Elevation    Final SH      Initial SH    Final SH

Newport River       Bottom     97.2 (3.5)       20.4       45.7 (0.5)
                    Low        98.0 (4.1)       20.7       50.8 (l.7)
                    High       91.7 (1.6)       20.4       57.4 (1.3)
Chadwick Bay        Bottom     91.1 (3.3)       20.1       52.4 (2.0)
                    Low        96.7 (l.3)       21.4       51.0 (2.2)
                    High       93.4 (1.4)       20.1       50.7 (1.7)

TABLE 4.

Prevalence and intensity of the oyster disease Perkinsus
marinus among native and non-native oysters grown at 7 field sites in
North Carolina. P. marinus prevalence and intensity were examined in
August and October of 2001.

                           Broad Creek   Swan Quarter    Bay River
Experimental Height            Low          Bay Low         Low

August 2001
C. arlakensis
  % infected                  0.0%           0.0%          0.0%
  Weighted intensity (a)      0.00           0.00          0.00
C. gigas
  % infected                                               5.6%
  Weighted intensity                                       0.00
C. virginica
  % infected                  0.0%           0.0%          0.0%
  Weighted intensity          0.00           0.00          0.00
October 2001
C. ariakensis
  % infected                  7.1%           0.0%          0.0%
  Weighted intensity          0.07           0.00          0.00
C. gigas
  % infected                                               0.0%
  Weighted intensity                                       0.00
C. virginica
  % infected                  13.3%          0.0%          6.7%
  Weighted intensity          0.13           0.00          0.07

                               Newport River        Chadwick Bay

Experimental Height        Bottom     Low    High   Bottom    Low

August 2001
C. arlakensis
  % infected                0.0%     0.0%    0.0%    0.0%     6.7%
  Weighted intensity (a)    0.00     0.00    0.00    0.00     0.07
C. gigas
  % infected                0.0%     0.0%    0.0%    0.0%     0.0%
  Weighted intensity        0.00     0.00    0.00    0.00     0.00
C. virginica
  % infected                0.0%     10.0%   0.0%    0.0%     0.0%
  Weighted intensity        0.00     0.10    0.00    0.00     0.00
October 2001
C. ariakensis
  % infected                0.0%     16.7%   0.0%    0.0%     0.0%
  Weighted intensity        0.00     0.17    0.00    0.00     0.00
C. gigas
  % infected                0.0%     0.0%    0.0%    0.0%     0.0%
  Weighted intensity        0.00     0.00    0.00    0.00     0.00
C. virginica
  % infected                0.0%     0.0%    0.0%    14.3%    0.0%
  Weighted intensity        0.00     0.00    0.00    0.14     0.00

                           Chadwick Bay
                                          Waters Bay   Topsail Sound
Experimental Height            High          Low            Low

August 2001
C. arlakensis
  % infected                   0.0%          6.7%          0.0%
  Weighted intensity (a)       0.00          0.07          0.00
C. gigas
  % infected                   0.0%          0.0%          0.0%
  Weighted intensity           0.00          0.00          0.00
C. virginica
  % infected                   0.0%          7.1%          0.0%
  Weighted intensity           0.00          0.07          0.00
October 2001
C. ariakensis
  % infected                  14.3%          0.0%          13.3%
  Weighted intensity           0.29          0.00           0.67
C. gigas
  % infected                   7.7%          0.0%           0.0%
  Weighted intensity           0.23          0.00           0.00
C. virginica
  % infected                   0.0%          7.7%          0.0%
  Weighted intensity           0.00          0.38          0.0%

(a) Infection intensity was calculated using the method described by
Ray (1954) and Mackin (1962), with infection inensity categorized into
the following groups: (0) absent, (1) light, (3) moderate, (5) heavy
(Calvo et al. 1999, Len ihan et al. 1999). Average weighted intenstiy
of dermo then was calculated for each species at each site by
multiplying the number of oysters with each infection level by its
infection intensity and dividing this sum by the total number of
oysters tested.

TABLE 5.

Prevalence and intensity of mud worm Polydora spp. infestation
among native and non-native oysters grown at 7 field sites in North
Carolina. Mud worm infestation levels were quantified in August and
October of 2001.

Experimental Height      Broad Creek   Swan Quarter    Bay River
                             Low          Bay Low         Low
August 2001
C. ariakensis
% W/Polydora               100.0%          86.7%         0.0%
Weighted intensity           3.00           1.40         0.00
C. gigas
% w/Polydora                                             0.0%
Weighted intensity (a)                                   0.00
C. virginica
% w/Polydora                0.0%           0.0%          0.0%
Weighted intensity          0.00           0.00          0.00
October 2001
C. ariakensis
% w/Polydora                28.6%          46.2%         60.0%
Weighted intensity           0.93           0.92          1.53
C. gigas
% w/Polydora                                             50.0%
Weighted intensity                                        1.60
C. virginica
% w/Polydora                0.0%           0.0%          0.0%
Weighted intensity          0.00           0.00          0.00

                              Newport River          Chadwick Bay

Experimental Height      Bottom     Low     High   Bottom      Low

August 2001
C. ariakensis
% W/Polydora              0.0%     0.0%    0.0%     33.3%     26.7%
Weighted intensity        0.00     0.00    0.00      0.89      0.33
C. gigas
% w/Polydora              0.0%     0.0%    0.0%     66.7%     40.0%
Weighted intensity (a)    0.00     0.00    0.00      1.27      0.80
C. virginica
% w/Polydora              0.0%     0.0%    0.0%      0.0%      0.0%
Weighted intensity        0.00     0.00    0.00      0.00      0.00
October 2001
C. ariakensis
% w/Polydora              0.0%     16.7%   28.6%    92.3%     100.0%
Weighted intensity        0.00      0.17    0.29     1.31       3.00
C. gigas
% w/Polydora              13.3%    53.8%   0.0%     66.7%     100.0%
Weighted intensity        0.13      0.77   0.00      1.67       2.00
C. virginica
% w/Polydora              0.0%     0.0%    0.0%      9.0%      0.0%
Weighted intensity        0.00     0.00    0.00      0.00      0.00

                         Chadwick Bay
                                        Waters Bay   Topsail Sound
Experimental Height          High          Low            Low

August 2001
C. ariakensis
% W/Polydora                23.8%          0.0%          0.0%
Weighted intensity           0.48          0.00          0.00
C. gigas
% w/Polydora                66.7%          6.7%          0.0%
Weighted intensity (a)       1.27          0.07          0.00
C. virginica
% w/Polydora                 0.0%          7.1%          0.0%
Weighted intensity           0.00          0.07          0.00
October 2001
C. ariakensis
% w/Polydora                64.3%          0.0%          0.0%
Weighted intensity           1.79          0.00          0.00
C. gigas
% w/Polydora                69.2%          0.0%          0.0%
Weighted intensity           1.54          0.00          0.00
C. virginica
% w/Polydora                 9.1%          0.0%          0.0%
Weighted intensity           0.09          0.00          0.00

Intensity of Polrdora spp. shell infestation was rated on a scale of 0
to 4 to approximate the percentage of the oyster shell covered by mud
worm tubes (101 absent, [1]<25%, [2] 25-509. [3] 50-75%. and [4] >75%
coverage: Calvo et al. 1999). Weighted intensity Has calculated by
multiplying the number of oysters with each intensity level by their
respective infestation intensity and dividing the sum by the total
number of oysters tested.

TABLE 6.

Summary of results from experimental culture of native (C.
virginica) versus non-native (C. ariakensis and C. gigas) during
1999-2000 and 2001-2002 in North Carolina.

                            Year        Site
1. High-salinity (>25%,)
sites                       1999-2000   Chadwick Bay

                            1999-2000   Chadwick Bay

                            1999-2000   Waters Bay

                            2001-2002   Waters Bay
                            2001-2002   Topsail Sound
Elevation experiment
                            2001-2002   Newport River
                            2001-2002   Newport River
                            2001-2002   Newport River
                            2001-2002   Chadwick Bay
                            2001-2002   Chadwick Bay
                            2001-2002   Chadwick Bay
                            2001-2002   Newport River
                            2001-2002   Newport River
                            2001-2002   Newport River
                            2001-2002   Chadwick Bay
                            2001-2002   Chadwick Bay
                            2001-2002   Chadwick Bay
2. Intermediate-salinity
(15-25[per thousand])
sites                       2001-2002   Broad Creek,
                                          Wanchese
                            2001-2002   Swan Quarter
                                          Bay
                            2001-2002   Bay River
3. Low-salinity
(<10[per thousand]) site    1999-2000   Broad Creek,
(b)                                       Wanchese

                             Species Compared        Elevation
1. High-salinity (>25%,)
sites                       C. ariakensis versus   Low elevation
                              C. virginica
                            C. gigas versus        Low elevation
                              C. virginica
                            C. gigas versus        Low elevation
                              C. virginica
                            All 3 species          Low elevation
                            All 3 species          Low elevation
Elevation experiment
                            All 3 species          Bottom
                            All 3 species          Low elevation
                            All 3 species          High elevation
                            All 3 species          Bottom
                            All 3 species          Low elevation
                            All 3 species          High elevation
                            C. ariakensis          All 3 elevations
                            C. gigas               All 3 elevations
                            C. virginica           All 3 elevations
                            C. ariakensis          All 3 elevations
                            C. gigas               All 3 elevations
                            C. virginica           All 3 elevations
2. Intermediate-salinity
(15-25[per thousand])
sites                       C. ariakensis versus   Low elevation
                              C. virginica
                            C. ariakensis versus   Low elevation
                              C. virginica
                            All 3 species          Low elevation
3. Low-salinity
(<10[per thousand]) site    C. ariakensis versus   Low elevation
(b)                           C. virginica

                                          Results

                            Individual Growth      Survivorship

1. High-salinity (>25%,)
sites                       aria > virg           virg > aria (a)

                            gigas > virg          giga > virg

                            gigas > virg          gigas > virg

                            gigas > aria > virg   aria = gigas > virg
                            gigas > aria > virg   aria = gigas > virg
Elevation experiment
                            gigas > aria > virg   gigas > aria = virg
                            gigas > aria > virg   gigas > virg > aria
                            gigas > aria > virg   gigas > aria > virg
                            gigas > aria > virg   gigas > aria = virg
                            gigas > aria > virg   gigas > aria = virg
                            gigas > aria > virg   gigas > aria > virg
                            high = low > bottom   high > low = bottom
                            bottom = low = high   bottom = low = high
                            high > low > bottom   high > bottom
                            high > low = bottom   high > low = bottom
                            bottom = low = high   bottom = low = high
                            bottom = low = high   high > bottom
2. Intermediate-salinity
(15-25[per thousand])
sites                       aria = virg           aria = virg

                            aria > virg           aria = virg

                            aria > virg > gigas   aria = virg > gigas
3. Low-salinity
(<10[per thousand]) site    virg > aria           virg > aria
(b)

(a) C. virginia but not reached marketable size by the end of the
experiment. Therefore, the 2-9 months of additional estimated
grow-out would be expected to lead to much more mortality from demo
and other sources.

(b) Abnormal environmental conditions following Hurricane Floyd
resulted in extremely low salinity levels at Broad Creek in 2000
(Peterson 2000).

ACKNOWLEDGMENTS

The authors thank Matt Kimble, Chris Stewart, Christina Tallent, and Rachael Wagaman for assistance in culturing the oysters in the field and conducting laboratory assays. Stan Allen, Jr., of the Virginia Institute of Marine Sciences provided disease-free triploid seed for both non-native oysters and much guidance. The authors thank Joe McClees, Carla Gwaltney, and numerous other supporters. This manuscript was greatly enhanced from comments by Sandy Shumway and two anonymous reviewers. Mack Salter and Brian Sheppard also contributed their knowledge as commercial shellfishermen. This research was supported by the North Carolina General Assembly (through the Rural Development Foundation and the Fishery Development Foundation) and the North Carolina Department of Environment and Natural Resources.

LITERATURE CITED 

Allen, S. K. & S. L. Downing. 1986. Performance of triploid Pacific oysters, Crassostrea Gigas (Thunberg). 1. Survival, growth, glycogen content, and sexual maturation in yearlings. J. Exper. Mar. Biol. Ecol. 102:197-208.

Andrews, J. D. 1980. A review of introductions of exotic oysters and biological planning for new importations. Mar. Fish. Rev. 42:1-11.

Ayers, P. 1991. Introduced Pacific oysters in Australia. In: R. Osman, ed. The ecology of Crassostrea gigas in Australia, New Zealand, France, and Washington State. College Park, MD: Maryland Sea Grant. pp. 3-7.

Barber, B. J. & R. Mann. 1994. Growth and mortality of eastern oysters, Crassostrea virginica (Gmelin, 1791), and Pacific oysters Crassostrea gigas (Thunberg, 1793) under challenge from the parasite Perkinsus marinus. J. Shellfish Res. 13:109-114.

Breese, W. P. & R. E. Malouf. 1977. Hatchery rearing techniques for the oyster Crassostrea rivularis Gould. Aquaculture 12:123-126.

Byrne, R. J. 1996. Strategic plan for molluscan shellfish research; including a rational plan for testing application of non-native oyster species. Report to the Governor and General Assembly of Virginia. House Document No. 16. Richmond, VA: Virginia Institute of Marine Sciences Publication. 123 pp.

Calvo, G. W., M. W. Luckenbach, S. K. Allen, Jr. & E. M. Burreson. 1999. Comparative field study of Crassostrea gigas (Thunberg, 1793) and Crassostrea virginica (Gmelin, 1791) in relation to salinity in Virginia. J. Shellfish Res. 18:465-473.

Calvo, G. W., M. W. Luckenbach, S. K. Allen, Jr. & E. M. Burreson. 2001. A comparative field study of Crassostrea ariakensis (Fujita 1913) and Crassostrea virginica (Gmelin 1791) in relation to salinity in Virginia. J. Shellfish Res. 20:221-229.

Carlton, J. T. 1992. Introduced marine and estuarine mollusks of North America: An end-of-the-20th-century perspective. J. Shellfish Res. 11: 489-505.

Carriker, M. R. 1992. Introductions and transfers of molluscs: Risk considerations. J. Shellfish Res. 11:507-510.

Courtney, W. R. J. & C. R. Robins. 1989. Fish introductions: good management, mismanagement, or no management? Aquatic Sci. 1:159-172.

Davis, J. P. 1989. Growth rate of sibling diploid and triploid oysters, Crassostrea gigas. J. Shellfish Res. 8:319.

Day, R. W. & G. P. Quinn. 1989. Comparisons of treatments after an analysis of variance in ecology. Ecol. Monographs 59:433-463.

Gaffney, P. M. & S. K. J. Allen. 1992. Genetic aspects of introduction and transfer of molluscs. J. Shellfish Res. 11:535-538.

Gaffney, P. M., V. P. Rubin, D. Hedgecock, D. A. Powers, G. Morris & L. Hereford. 1996. Genetic effects of artificial propagation: Signals from wild and hatchery populations of red abalone in California. Aquaculture 143:257-266.

Gottlieb, S. J. & M. E. Schweighofer. 1996. Oysters and the Chesapeake Bay ecosystem: A case for exotic species introduction to improve environmental quality? Estuaries 19:639-650.

Grabowski, J. H., S. P. Powers, C. H. Peterson, M. J. Powers & D. Green. 2003. Consumer ratings of non-native (Crassostrea gigas and Crassostrea ariakensis) vs. native (Crassostrea virginica) oysters. J. Shellfish Res. 22:21-30.

Langdon, C. J. & A. M. Robinson. 1996. Aquaculture potential of the Suminoe oyster (Crassostrea ariakensis Fugita 1913). Aquaculture 144:321-338.

Launey, S. & D. Hedgecock. 2001. High genetic load in the Pacific oyster Crassostrea gigas. Genetics 159:255-265.

Lenihan, H. S., F. Micheli, S. W. Shelton & C. H. Peterson. 1999. The influence of multiple environmental stressors on susceptibility to parasites: An experimental determination with oysters. Limnol. Oceanography 44:910-924.

Lipton, D. W., E. F. Lavan & I. E. Strand. 1992. Economics of molluscan introductions and transfers: The Chesapeake Bay dilemma. J. Shellfish Res. 11:511-519.

Mackin, J. G. 1962. Oyster disease caused by Dermocystidium marinum. Publication of the Institute of Marine Sciences, University of Texas 7:132-229.

Mann, R., E. M. Burreson & P. K. Baker. 1991. The decline of the Virginia oyster fishery in Chesapeake Bay: Considerations for introduction of a non-endemic species, Crassostrea gigas (Thunberg, 1973). J. Shellfish Res. 10:379-388.

Mann, R. E. 1979. Exotic Species in Mariculture. Cambridge, Massachusetts: MIT Press. 363 pp.

National Research Council. 2003. Non-native Oysters in the Chesapeake Bay. Committee on Non-native Oysters in the Chesapeake Bay, Ocean Studies Board, Division on Earth and Life Studies. Washington, DC: The National Academies Press.

Paynter, K. T. & E. M. Burreson. 1991. Effects of Perkinsus marinus infection on the eastern oyster, Crassostrea virginica: 2. disease development and impact on growth rate at different salinities. J. Shellfish Res. 10:425-431.

Peterson, C. H. 2000. Sustainable oyster aquaculture study. Institute of Marine Sciences, University of North Carolina at Chapel Hill, Morehead City. 17 pp.

Ray, S. M. 1952. A culture technique for diagnosis of infections with Dermocystidium marinum. Science 116:360-361.

Ray, S. M. 1954. Biological studies of Dermocystidium marinum. The Rice Institute Pamphlet. Special Issue. Houston, TX. 111 pp.

Ray, S. M. 1963. A review of the culture method for detecting D. marinum, with suggested modifications and precautions. Proc.Nat. Shellfish. Assoc. 54:389-404.

Rhoads, D. C. & D. K. Young. 1970. The influence of deposit-feeding organisms on sediment stability and community trophic structure. J. Mar. Res.28:150-178.

Robinson, A. M. & C. J. Langdon. 1993. The Suminoe oyster-candidate for the half-shell trade? J. Shellfish Res. 12: 152.

Rosenfield, A. & F. G. Kern. 1979. Molluscan imports and the potential for introduction of disease organisms. In: R. Mann, editor. Exotic species in mariculture. Cambridge, Massachusetts: MIT Press. pp. 165-189.

Ruiz, G. M., P. W. Fofonoff, J. T. Carlton, M. J. Wonham & A. H. Hines. 2000. Invasion of coastal marine communities in North America: Apparent patterns, processes, and biases. Ann. Rev. Ecol. System. 31:481-531.

Shatkin, G., S. E. Shumway & R. Hawes. 1997. Considerations regarding the possible introduction of the Pacific Oyster (Crassostrea gigas) to the Gulf of Maine: a review of global experience. J. Shellfish Res. 16:463-477.

Underwood, A. J. 1981. Techniques of analysis of variance in experimental marine biology and ecology. Oceanogr. Mar. Biol. Annu. Rev. 19:513-605.

JONATHAN H. GRABOWSKI, * CHARLES H. PETERSON, SEAN P. POWERS, DAVID GASKILL AND HENRY C. SUMMERSON

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

University of North Carolina at Chapel Hill, Institute of Marine Sciences, Morehead City, North Carolina 28557