Naturalized populations of oysters, Crassostrea gigas along the South African coast: distribution, abundance and population structure.

ABSTRACT The Japanese oyster Crassostrea gigas forms the mainstay of the South African oyster farming industry. Despite having been cultured along the South African coast for 30 years, this is the first record of naturalized populations of this well known global invader from the region. DNA sequence data confirmed the presence of C. gigas in the Breede and Goukou estuaries. Populations of 184,206 [+ or -] 21 058.9 (SE), 876 [+ or -] 604.2 (SE) and 1 228 [+ or -] 841.8 (S.E.) individuals were recorded in the Breede, Goukou and Knysna estuaries respectively. No C. gigas were recorded from the open coast. The condition index of oysters in the Goukou and Knysna estuaries differed significantly from those in the Breede Estuary (Kruskal-Wallis ANOVA H = 48.249, P < 0.01) but no difference was found between the Goukou and Knysna estuaries (Kruskal-Wallis ANOVA H = 48.249, P > 0.05). The lower condition index recorded in the Breede Estuary is believed to reflect decreased food supply as a result of the large C. gigas population present in the estuary.

KEY WORDS: Crassostrea gigas, oyster, South Africa, introduced species, diagnostic DNA identification

INTRODUCTION

Over the last century, oysters have been extensively moved around the globe as highly successful aquaculture species (Andrews 1980). As the earliest of these translocations took place before the implications of introducing non-native species were realized or understood, they very often lead to the establishment of large invasive populations. Despite a more responsible approach to translocations of nonindigenous species in recent times (Shatkin et al. 1997), modern introductions have none-the-less resulted in naturalized populations establishing. Of all oysters used in aquaculture, the Japanese oyster Crassostrea gigas Thunberg is arguably the most aggressive invader. Although originating in the north west Pacific, C. gigas presently has established naturalized populations on all major coastlines of the Northern Hemisphere (with the exception of the Atlantic coast of North America) (Andrews 1980, Carlton et a1. 1990, Minchin 1996, Shatkin et al. 1997, Wehrmann et al. 2000), as well as on the coasts of Tasmania, southern and eastern Australia (Thomson 1959, Shatkin et al. 1997) and New Zealand (Dinamani 1971). It is however unclear if C. gigas exhibits heightened invasiveness in comparison to other common aquaculture species, or if the high frequency of transport of this species is responsible for the elevated incidence of naturalized populations.

Following global trends, the South African oyster industry is based on C. gigas. This species was first imported to Knysna Estuary in the 1950s (Korringa 1956), where it is still farmed. Presently two other South African estuaries (Kowie & Swartkops Estuaries) support oyster farms, whereas a further three fully marine farms are based in Algoa Bay, Saldanha Bay and Alexander Bay (Fig. 1). Due to difficulties in inducing predictable spawning and subsequent settling under South African conditions, the entire industry has been based on spat imported from Chile, the United Kingdom and France. As C. gigas has appeared unable to complete its life cycle under local environmental conditions, this species has not previously been considered to pose a threat as an invader along South African shores (Griffiths et al. 1992, C. Smith pers comm.).

[FIGURE 1 OMITTED]

This article documents the first record of naturalized populations of C. gigas along the coast of South Africa, and establishes the present distribution and status of this species.

METHODS

In 2001, 20 estuaries along the southern and eastern coasts of South Africa were surveyed for the presence of oysters. In instances where the identification of individuals was not possible using morphometric characters, tissue samples were collected and later used in genetic analyses to obtain confirmation of species identification.

Several recent studies have used DNA sequences of either mitochondrial protein coding, ribosomal genes (rDNA) or nuclear rDNA genes for studying phylogenetic relationships among oysters (Jozefowicz and Foighil 1998, Lapegue et al. 2002, Matsumoto 2003) as well as for species identification purposes (Banks et al. 1993). In this study C. gigas (n = 18) from 3 oyster farms (Walvis Bay and Swakopmund in Namibia and Saldanha Bay in South Africa) were used to compare with the indigenous Cape rock oyster Striostrea margaritacea Lamarck (n = 9), collected from the Breede (samples 1-6) and Goukou (samples 7-9) Estuaries, Diagnostic genetic differences were used to determine the identity of 12 C. gigas specimens collected from the Breede Estuary. Several published sequences were used for comparative purposes (numbers in parentheses refer to sequence accession numbers in a range of international data bases such as GenBank): C. gigas (COI, GB AF152565; 18S rDNA DBJ AB064942.1), Crassostrea virginica Gmelin (COI, GB AF152566; 18S rDNA EMB X60315.1), Crassostrea angulata Lamarck (COI, GB AF152567) and Saccostrea cuccullata Born (18S rDNA, EMB AJ389634.1). Ostrea chilensis_Philippi (COI, GB AF112289) or O. edulis (18S rDNA, GB U88709.1) was used for outgroup rooting of phylogenetic trees.

DNA was extracted from 2 mg of gill tissue using the DNeasy 96 Tissue Kit (Qiagen) according to the manufacturer's instructions. Two DNA regions were amplified via the polymerase chain reaction (PCR; Saiki et al. 1988): A 700 base pair (bp) fragment of the mitochondrial DNA (mtDNA) cytochrome oxidase subunit I (COI) gene (Folmer et al. 1994) and a 520 bp fragment of the 18S nuclear rDNA gene (Lane et al. 1985, Diez et al. 2001). Amplifications were performed in a total volume of 25 [micro]l containing 0.5 units Taq DNA polymerase (Invitrogen). For COI 2.5 mM Mg[Cl.sub.2] was used at an annealing temperature of 45[degrees]C, while 18S was annealed at 50[degrees]C in reactions containing 1.5 mM Mg[Cl.sub.2]. Successful PCR reactions were purified using the SAP/Exo system according to the manufacturer's protocol (http://www.pebiodocs. com/pebiodocs/04305080.pdf). Successful PCR reactions were sequenced in the forward and reverse directions using dye-terminator cycle sequencing (Big Dye DNA sequencing kit, Applied Biosystems) with enzymatic template purification according to the manufacturer's recommendations; sequences were electrophoresed on an ABI 3100 automated capillary DNA sequencer (Applied Bitsystems). Sequences were deposited in GenBank.

Consensus sequences of each specimen were aligned to each other in ClustalX (Thompson et al. 1997). Aligned sequences were compared in MEGA version 2.1 (Kumar et al. 2001) where genetic distances were calculated according to the Kimura-two-parameter model (Kimura 1980) and sequences were clustered using the neighbor-joining algorithm (Saitou & Nei 1987). Confidence in nodes was assessed using bootstrapping (Felsenstein 1985). Sequence statistics such as nucleotide diversity (Saitou & Nei 1987) was also calculated in MEGA. A diagnostic restriction enzyme test was developed based on observed sequence differences.

Following genetic confirmation of the presence of C. gigas within South African waters, the six estuaries in which this species was originally found were extensively resampled in 2003. During resampling, the morphologic characteristics detailed in Table 1 were used to distinguish C. gigas from indigenous oyster species. In each estuary, all rocky shores and artificial structures offering suitable habitat for C. gigas were visited, and at least four intertidal transects were surveyed for every 200 m of suitable habitat. At each transect site the intertidal was divided into three tidal zones (i.e., high-, mid- and low-shore zones). Using a 0.5 [m.sub.2] quadrat, which was rolled up the shore from Mean Low Water Spring to Mean High Water Spring, the number of C. gigas per quadrat, in each tidal zone, was recorded. This was used to calculate the mean number of C. gigas individuals supported per meter square in the three tidal zones as well as the total population present in each estuary. Due to the eco-morphic nature of oysters, the capacity of the inner side of the shell was used as a measure of size of individuals. Sample sizes varied between the estuaries depending on the size of the population. Individuals were randomly selected from various tidal heights and at varying distances from the mouth of the estuary. Each individual was weighed whole to the nearest 0.01g. The shell and soft tissue were then separated. Shells were dried at room temperature for 24 h and then weighed again, whereas the soft tissues were oven dried at 105[degrees]C for 24 h and weighed. The following equation was used when calculating inner shell capacity (Cosby & Gale 1990):

Capacity of the inner side of the shell = Whole wet weight (g)--Dry shell weight (g)

Condition index (CI) was calculated on a gravimetric basis using the following equation (Rainer & Mann 1992):

CI = Soft tissue dry weight (g) x 1,000 Capacity of the inner shell (g)

Besides sampling the estuaries in which C. gigas was first recorded in 2001, three sites along the open coast between each of the estuaries were surveyed for the presence of C. gigas during 2003 (Fig. 1).

Statistical Analyses

The sizes of C. gigas individuals occurring in the three estuaries were compared using the Kruskal-Wallis ANOVA for multiple comparisons as non-normal distributions were recorded (Kolmogorov-Smirnov one-sample test). Because the condition indices of individuals were found to have heteroskedastic distributions (Kolmogorov-Smirnov one-sample test), they were compared using the Kruskal-Wallis ANOVA, followed by nonparametric multiple comparisons. All statistical analyses were conducted using STATISTICA for Windows (Version 6), StatSoft Inc. (2001).

RESULTS

Indigenous or introduced oyster populations were recorded in all but one of the estuaries (Keurbooms Estuary) surveyed during 2001 (Table 2). Of the indigenous species S. margaritacea was found throughout the survey region, whereas estuaries to the east of Keiskamma also harbored populations of the Natal rock oyster S. cucculata. Natural populations of oysters appearing unlike either known indigenous oyster species and morphologically resembling C. gigas were discovered in six estuaries in the region (Table 2). This identification was subsequently genetically confirmed.

DNA sequences of the mtDNA COI gene showed substantial differences between C. gigas and S. margaritacea (Fig. 2a). These sequences in comparison with other published oyster sequences also revealed very high levels of sequence divergence between these taxa and S. margaritacea (109-140 nucleotide differences or 27% to 36% nucleotide sequence divergence and 21-30 amino acid differences). When compared with other species based on nucleotide divergence, Crassostrea species grouped with C. gigas to the exclusion of S. margaritacea (Fig. 3; S. cucculata COI sequence was unavailable). However, four exceptional S. margaritacea samples (samples 3, 7-9) grouped with C. gigas suggesting either misidentification or hybridization. Mitochondrial DNA is haploid and therefore cannot be used to test for hybridization. Multicopy nuclear DNA loci such as rRNA are expected to show, within an individual, a mixture of polymorphism from two species that recently hybridized. Nuclear 18S sequences showed a large divergence between C. gigas and S. margaritacea (Fig. 2b) and grouped S. margaritacea separately from Crassostrea and Saccostrea species (Fig. 4). The four exceptional S. margaritacea samples with C. gigas mtDNA, had 18S polymorphisms characteristic of C. gigas and there is thus no genetic evidence of hybridization.

[FIGURES 2-4 OMITTED]

A diagnostic restriction pattern was found between C. gigas and S. margaritacea COI using the HaeIII restriction, enzyme that recognizes the DNA sequence 5' GG[down arrow]CC 3' (Fig, 2a). To confirm identifications, all amplified COI products (n = 39) were subsequently digested with HaeIII to identify S. margaritacea based on the absence of a cut and C. gigas based on a two-fragment electrophoretic profile. All C. gigas from the oyster farms showed the C. gigas restriction profile. All S. margaritacea restriction patterns confirmed the earlier sequencing results as outlined earlier. The 12 specimens from the Breede estuary that were morphologically identified as C. gigas, showed the C. gigas restriction profile (seven of the samples were also sequenced, see Fig. 3). Overall C. gigas from the oyster farms and from the estuaries showed very low levels of nucleotide diversity ([pi] = 0.001 or 0.1%). The local S. margaritacea showed slightly higher levels of nucleotide diversity ([pi] = 0.003 or 0.3%) although this estimate should be recalculated based on a larger number of individuals.

Crassostrea gigas populations have probably not been recognized along the South African coast to date because they are so easily confused with the two indigenous species found in the region. To simplify field identification, the physical features, which can be used to distinguish C. gigas from the two dominant indigenous intertidal oysters found along the South African coast are summarized in Table 1. No single feature provides unequivocal identification of these oysters, because natural variability is great. Rather, a combination of features is recommended to ensure accurate identification of all size classes. The most discerning feature of C. gigas is the presence of at least one purple adductor scar on a non-iridescent white shell interior. The vertical range of this species, from the low-shore zone to a depth of 1 m, is also an important field characteristic. The Cape rock oyster S. margaritacea is most reliably distinguished by the presence of fine radial threads on the surface of the top valve. In addition, an iridescent mother-of-pearl shell interior and smooth margins on both valves are useful identifying features. The natal rock oyster S. cuccullata is much smaller than either C. gigas or S. margaritacea, and is easily distinguished by the presence of a black adductor scar on the top valve, while no scar is visible on the lower valve. Blackened valve margins with undulating folds are also present on most individuals.

Although C. gigas was recorded in six estuaries along the southern and eastern coasts in 2001 (Table 2), wild populations were only present in three of these estuaries in 2003. These were the Breede, Goukou and Knysna Estuaries. No C. gigas were recorded on the open coast. In all three of these estuaries the entire population was restricted to the low-shore zone. Breede Estuary supported by far the highest density of individuals per meter square (i.e., 8.3 [+ or -] 8.8 [SD] * [m.sup.-2]) while the Goukou and Knysna Estuaries supported 1.0 [+ or -] 2.6 (SD) and 0.9 [+ or -] 2.1 (SD) individuals.[m.sup.-2] respectively (Fig. 5). Total population size within the Breede Estuary was 184 206 [+ or -] 21 058.9 (SE) individuals (95% confidence limits = 0-496 363), compared with only 876 [+ or -] 604.2 (SE) individuals (95% confidence limits = 0-5 482) and 1 228 [+ or -] 841.8 (SE) individuals (95% confidence limits = 0-6 614) in the Goukou and Knysna Estuaries respectively. Thus the Breede Estuary population is more than two orders of magnitude larger than those in either of the other two estuaries.

[FIGURE 5 OMITTED]

No significant difference was recorded in the size of individuals from the various estuaries (Kruskal-Wallis ANOVA H = 4.547, P > 0.05). The Goukou Estuary supported the largest individuals (40.95 [+ or -] 57.84 SD), followed by the Breede (37.02 [+ or -] 26.08) and Knysna (24.47 [+ or -] 19.34) Estuaries.

Although the smallest population of C. gigas was recorded in the Goukou Estuary, these oysters had a higher mean CI than those in the other estuaries (Fig. 6). The CI of C. gigas differed significantly between estuaries in the Goukou and Knysna Estuaries, from those in the Breede Estuary (Kruskal-Wallis ANOVA H = 48.249, P < 0.01), but no difference was found between the Goukou and Knysna Estuaries (Kruskal-Wallis ANOVA H = 48.249, P > 0.05).

[FIGURE 6 OMITTED]

DISCUSSION

There are numerous records from around the world of C. gigas being introduced as an aquaculture species, and remaining confined to these operations for many years; however, unpredictable recruitment and subsequent establishment of natural populations has eventually occurred in nearly all cases (Shatkin et al. 1997, Wehrmann et al. 2000). Examples of such "dormancy" have been shown in the Netherlands and in the United Kingdom, where C. gigas was first imported in 1964 and 1965, but only established wild populations in 1976 and 1991 respectively (Mann 1983, Shatkin et al. 1997). In both cases, unpredicted increases in sea temperature probably enabled the establishment of naturalized stocks. The history of C. gigas along South African shores has shown a similar pattern. The species was first introduced in the 1950s (Korringa 1956) and wild populations were first recorded only in 2001. The difference in the South African situation is that the southern and eastern coasts offer a coastal environment with sea temperatures varying between 10[degrees]C and 21[degrees]C (Lutjeharms 1998)--well within the range tolerated by C. gigas during the completion of its lifecycle (i.e., 4[degrees]C-24[degrees]C) (Shatkin et al. 1997). Thus, the delayed invasion of the South African coast by C. gigas has not been facilitated by a change in sea temperature and the trigger factor specific to this invasion remains unclear at this stage.

DNA sequence data of the mtDNA COI gene and the nuclear 18S rRNA gene confirmed the presence of C. gigas in the Breede estuary. In addition a few specimens from the Breede and Goukou estuaries, morphologically identified as S. margaritacea, were genetically identified as C. gigas. Based on the 18S rDNA results these specimens are unlikely to represent hybrids. In addition, Menzel (1987) reported no fertilization between C. virginica and S. margaritacea. Although no direct test of cross-fertilization exists for Cape rock oyster and C. gigas, based on the phylogenetic relationships found in our study (Figs. 3 and 4), morphologic plasticity represents a more likely explanation for the disparity between morphologic and genetic identification of the four specimens. The latter may introduce some uncertainty as to the extent and timeframe of naturalization of C. gigas in South African waters. Most morphologic identifications in the present study were however correct.

The presence of wild populations of C. gigas in the Knysna, Kromme and Keiskamma Estuaries (all of which supported oyster farms) in 2001, demonstrates the strong link between the spread of this species and oyster aquaculture operations, It is notable that despite the presence of oyster farms in the Kowie and Swartkops Estuaries, no C. gigas populations have established there to date. These estuaries should, however, be noted as potential locations for future invasions. The presence of C. gigas in the Breede (2001, 2003), Duiwenhoks (2001) and Goukou (2001, 2003) Estuaries (none of which have ever supported oyster farms), beg the question as to how this species managed to spread to these sites. Future studies combining morphology and molecular markers may clarify whether the various South African populations of C. gigas can be attributed to spread following a single introduction, or if in fact multiple introductions have occurred.

It is possible that C. gigas larvae were carried westwards from Knysna by the Agulhas current (Fig. 1) and were then taken into the estuaries (all of which are open throughout the year) during tidal exchange. Live adult oysters bought for personal consumption, may also have been deliberately, but illegally, introduced by waterfront property owners. A third possibility is that oysters purchased for personal consumption may have been stored in the estuaries for short periods prior to consumption and have spawned over this time. Holidaymakers frequent these areas particularly during the December and April holidays and C. gigas from Knysna (the most likely source of oysters for holidaymakers in this region) are known to spawn between September and April (Tonin pers. ob.). This makes it quite possible that spawning of farmed individuals temporarily held in the estuaries could have resulted in the establishment of the naturalized populations.

Knysna Estuary is the largest estuary along the southern coast. This, combined with the fact that it supports a large oyster farm, might lead one to expect that the largest wild population of C. gigas would occur within this estuary. In fact, Knysna Estuary supported a population of only 1 228 [+ or -] 841.8 (SE) Japanese Oysters (95% confidence limits = 0-6 614), while by far the largest population (184 206 [+ or -] 21 058.9 (SE) individuals (95% confidence limits = 0-496 363)) was recorded in the Breede Estuary. The reason for the observed pattern may relate to flooding of the rivers feeding the respective estuaries. Knysna Estuary experienced extensive and prolonged flooding in 1998, which bad severe effects on many intertidal invertebrates (P. Joubert pers comm.). In particular, it was noticed by South African National Parks officials that the number of oysters in the estuary declined dramatically after flooding subsided. Thus, the present size of the C. gigas population in the Knysna Estuary may reflect a population yet to recover from the effects of flooding.

The significantly higher CI of C. gigas individuals from the Goukou and Knysna estuaries, in comparison to the Breede Estuary, may reflect a limited food supply available to Breede individuals as a result of the dramatically larger population in this estuary.

The absence of C. gigas from open coasts adjoining invaded estuaries may reflect the mediating effect of wave action on invasive species recorded by other authors along the South African coast. The density, recruitment and growth of the Mediterranean mussel Mytilus galloprovincialis Lamarck along the west coast of South Africa, has been found to peak under conditions of high wave exposure while decreasing towards the two extremes wave energy (Steffani & Branch 2003a, 2003b, 2003c). Similarly, the European shore crab Carcinus maens Linnaeus has been shown to be restricted in abundance along open coast wave exposed shores while flourishing in areas of shelter (Le Roux et al. 1990, Robinson in prep). Should the high energy of the South African coast prove limiting to the distribution of C. gigas, the yet undetermined impact of this species may be focused on estuarine habitats and not on the coastal intertidal as has been reported globally (Shatkin et al. 1997).

As the presence of naturalized C. gigas along the South African coast has only just been discovered, the rate of spread of this species remains unknown, as do the ecologic implications of this invasion. Internationally, impacts resulting from the invasion of C. gigas include the simultaneous introduction of associated fauna (Critchley et al. 1983, Kaiser et al 1998), the introduction of disease organisms (Ford 1992), genetic pollution of local oyster species (Gaffney and Allen 1992, Gaffney and Allen 1993) and the reduction of indigenous oyster populations to threatened levels (Williams et al. 1988, Mann et al. 1991). It is thus clear, that the invasion of the South African coast by C. gigas requires urgent scientific attention if its environmental impacts are to be managed responsibly.

ACKNOWLEDGMENTS

The authors thank South African National Parks for facilitating sampling within the Knysna Estuary and Dr. Anesh Govender for providing valuable statistical assistance. Funding for this study was provided by the National Research Foundation and Marine and Coastal Management, through a grant to C L Griffiths from the Sea and the Coast 2 Programme. The genetic analyses were funded by a National Research Foundation and Marine and Coastal Management sabbatical grant to P Bloomer and a SeaGrant to M P Hare.

LITERATURE CITED

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

Banks, M. A., D. Hedgecock & C. Waters. 1993. Discrimination between closely related Pacific oyster spp. (Crassostrea) via mitochondrial DNA sequences encoding for large subunit rRNA. Mol. Mar. Biol. Biotechnol 2:129-136.

Carlton, J. T., J. K. Thompson, L. E. Schemel & F. H. Nichols. 1990. Remarkable invasion of San Francisco Bay (California, USA) by the Asian clam Potamocorbula amurensis. I. Introduction and dispersal. Mar. Ecol. Prog. Ser. 66:81-94.

Cosby, M.P. & L.D. Gale. 1990. A review and evaluation of bivalve condition index methodologies with a suggested standard method. J. Shellfish Res. 9:233-237.

Critchley, A. T., W. F. Farnham & S. L. Morrell. 1983. A chronology of new European sites of attachment for the invasive brown algae Sargassum muticum, 1973-1981. J. Mar. Biol. Assoc. UK. 63:799-811.

Diez, B., C. Pedros-Alio, T. L. Marsh & R. Massana. 2001. Application of denaturing gradient gel electrophoresis (DGGE) to study the diversity of marine picoeukaryotic assemblages and comparison of DGGE with other molecular techniques. Appl. Env. Microbiol. 67:2942-2951.

Dinamani, P. 1971. Occurrence of the Japanese oyster, Crassostrea gigas (Thunberg), in Northland, New Zealand. New Zealand J. Mar. Freshwater Res. 5:352-357.

Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791.

Ford, S.E. 1992. Avoiding the transmission of disease in commercial culture of mollusks, with special reference to Perkinsus marinus (Dermo) and Haplosporidium nelsoni (MSX). J. Shellfish Res. 11:539-546.

Folmer, O., M. Black, W. Hoeh, R. Lutz & R. Vrijenhoek. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotech. 3:294-299.

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

Gaffney, P.M. & S. K. Allen. 1993. Hybridization among Crassostrea species: a review. Aquaculture 116:1-13.

Griffiths, C.L., P. A. R. Hockey, C. Van Erkom Schurink, & P.J. Le Roux. 1992. Marine invasive aliens on South African shores: Implications for community structure and trophic functioning. S. Afr. J. Mar Sci 12:713-722.

Jozefowicz, C. J. & D. O. Foighil. 1998. Phylogenetic analysis of Southern Hemisphere flat oysters based on partial mitochondrial 16S rDNA gene sequences. Mol. Phylogenet. Evol. 10:426-435.

Kaiser, M. J., I. Laing, D. S. Utting & G. M. Burnell. 1998. Environmental impacts of bivalve mariculture. J. Shellfish Res. 17:59-66.

Kimura, M. 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16:111-120.

Korringa, P. 1956. Oyster culture in South Africa. Hydrographical, biological and ostreological observations in the Knysna Lagoon, with notes on conditions in other South African waters. Invest. Rep. Div. Fish. S. Afr. 20:84.

Kumar, T., K. Tamura, I. Jakobsen, & M. Nei. 2001. MEGA2: Molecular Evolutionary Genetic Analysis software. Bioinformatics 17:1244-1245.

Lane, D. J., B. Pace, G. J. Olsen, D. A. Stahl, M. L. Sogin & N. R. Pace. 1985. Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc. Natl. Acad. Sci. USA 82:6955-6959.

Lapegue, S., I. Boutet, A. Leitao, S. Heurtebise, P. Garcia, C. Thiriot-Quievreux & P. Boudry. 2002. Trans-Atlantic distribution of a mangrove oyster species revealed by 16S mtDNA and karyological analyses. Biol. Bull. 202:232-242.

Le Roux, P. J., G. M. Branch & M. A. P. Joska. 1990. On the distribution, diet and possible impact of the invasive European shore crab Carcinus maens (L.) along the South African Coast. S. Afr. Z Mar. Sci. 9:85-93.

Lutjeharms, J. R. E. 1998. Coastal hydrography. In: R. Lubke & I. Moor, editors. Field guide to the Eastern and Southern Cape coasts. Cape Town, South Africa: University of Cape Town Press. pp. 50-61.

Mann, R., E. Burreson & P. Baker. 1991. The decline of the Virginia oyster fishery in Chesapeake Bay: considerations for introduction of a nonendemic species, Crassostrea gigas (Thunberg 1793). J. Shellfish Res. 10:379-388.

Mann, R. 1983. The role of introduced bivalve mollusk species in mariculture. J. World Mariculture Soc. 14:546-559.

Matsumoto, M. 2003. Phylogenetic analysis of the subclass Pteriomorpha (Bivalvia) from mtDNA COI sequences. Mol. Phylogenet. Evol. 27: 429-440.

Menzel, W. 1987. Hybridization of oysters and clams. In: K. Tiews, editor. Selection, hybridization, and genetic engineering in Aquaculture. Vol. 2. Berlin:Heeneman. pp. 47-59.

Minchin, D. 1996. Management of the introduction and transfer of marine mollusks. Aquatic Conservation. Mar. Freshwater Eco. 6:229-244.

Rainer, J. S. & R. Mann. 1992. A comparison of methods for calculating condition index in Eastern Oysters, Crassostrea virginica (Gmelin, 1791). J. Shellfish Res. 11:55-58.

Robinson, T. B., C. L. Griffiths, C. D. McQuaid, & M. Ruis. In Press. Marine alien species of South Africa--status and impacts. Afr. J. Mar Sci.

Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K. B. Mullis & H. A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with a thermostabile DNA polymerase. Science 239: 487-491.

Saitou, N. & M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.

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.

Steffani, C. N. & G. M. Branch. 2003a. Growth rate, condition, and shell shape of Mytilus galloprovincialis: responses to wave exposure. Mar. Ecol. Prog. Ser. 246:197-209.

Steffani, C. N. & G. M. Branch. 2003b. Spatial comparisons of populations of an indigenous limpet Scutellastra argenvillei and the alien mussel Mytilus galloprovincialis along a gradient of wave energy. S. Afr. J. Mar. Sci. 25:195-212.

Steffani, C. N. & G. M. Branch. 2003c. Temporal changes in an interaction between an indigenous limpet Scutellastra argenvillei and an alien mussel Mytilus galloprovincialis: effects of wave exposure. S. Afr. J. Mar. Sci. 25:213-229.

Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin & D. G. Higgins. 1997. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl. Acids Res. 24:4876-4882.

Thomson, J. 1959. The naturalization of the Pacific oyster in Australia. Aust. J. Mar. Freshwater Res. 10:144-149.

Wehrmann, A., M. Herlyn, F. Bungenstock, G. Hertweck & G. Millat. 2000. The distribution gap is closed--first record of naturally settled Pacific oysters Crassostrea gigas in the east Frisian Wadden Sea, North Sea. Senckenbergiana Marit. 30:153-160.

Williams, R. J., F. B. Griffiths, E. J. Van Der Wal, & J. Kelly. 1988. Cargo vessel ballast water as a vector for the transport of non-indigenous marine species. Estuarine, Coastal and Shelf Science 27:409-420.

T. B. ROBINSON, (1) C. L. GRIFFITHS, (1) A. TONIN, (2) P. BLOOMER (3) AND M. P. HARE (4)

(1) Centre of Invasion Biology, Zoology Department and Marine Biology Research Institute, University of Cape Town, Rondebosch 7700, South Africa; (2) Mariculture Development Services, PO Box 200, Alexander Bay 8290; (3) Molecular Ecology and Evolution Programme, Department of Genetics, University of Pretoria, 0002, South Africa; (4) Biology Department, University of Maryland, College Park, Maryland 20742

TABLE 1.
Characterising features of the three dominant intertidal
oyster species found along the South African coast.

                    Crassostrea gigas          Striostrea margaritacea

Radial threads      No radical threads         Fine radial threads on
                                                 the surface of the
                                                 top valve
Shell interior      Non-iridescent: White to   Iridescent: Mother-of-
                      off-white, may have        pearl to gold in
                      mauve flecks               color
Adductor scar       At least one adductor      Most scars have no
                      scar purple in color       distinct color, some
                                                 may be tinged
                                                 pink-purple
Valve margins       Generally show no          Margins smooth with no
                      coloration but few         distinct coloration
                      may be mauve-black.
                      Margins most often
                      undulating
Max. size           200 mm                     180 mm
Location            Select estuaries on the    Mozambique to False
                      south coast. Not known     Bay. Predominantly
                      from the open coast        on the open coast, but
                      along SA shores            penetrate the mouths
                                                 of estuaries
Vertical position   Low intertidal zone to     Extreme low tide to a
  on shore            a depth of 1 m             depth of 5 m
Diagram             [ILLUSTRATION OMITTED]     [ILLUSTRATION OMITTED]

                    Saccostrea cuceullata

Radial threads      No radial threads
Shell interior      Non-iridescent: White to grey
Adductor scar       Adductor scar on top valve always black
                      in color, lower scar not coloured
Valve margins       Margins of both valves generally black
                      with undulating folds
Max. size           70 mm
Location            Indo-West-Pacific to Algoa Bay. Found
                      mainly on the open coast but may
                      occur in estuaries
Vertical position   Upper mid-intertidal zone
  on shore
Diagram             [ILLUSTRATION OMITTED]

TABLE 2.
A list of all estuaries searched for natural
populations of oysters in 2001. The presence
of C. gigas populations are indicated in bold.

                                                    Oyster
                                                    Species
                                                    Present

                         Geographical             Crassostrea
   Estuary                 Position                  gigas

1. Breede       34[degrees]26'S 20[degrees]52'E     [check]
2. Duiwenhoks   34[degrees]24'S 2l[degrees]00'E     [check]
3. Goukou       34[degrees]26'S 21[degrees]25'E     [check]
4. Gourits      34[degrees]24'S 22[degrees]82'E
5. Knysna       34[degrees]09'S 23[degrees]04'E     [check]
6. Keurbooms    34[degrees]01'S 23[degrees]23'E
7. Kromme       34[degrees]10'S 24[degrees]53'E     [check]
8. Swartkops    33[degrees]56'S 25[degrees]38'E
9. Sundays      33[degrees]45'S 25[degrees]52'E
10. Bushmans    33[degrees]44'S 26[degrees]40'E
11. Kariega     33[degrees]43'S 26[degrees]41'E
12. Kowie       33[degrees]37'S 36[degrees]55'E
13. Keiskamma   33[degrees]20'S 27[degrees]29'E     [check]
14. Tyolomnqa   33[degrees]16'S 27[degrees]33'E
15. Kwelera     32[degrees]59'S 28[degrees]03'E
16. Qora        32[degrees]31'S 28[degrees]41'E
17. Mbashe      32[degrees]17'S 28[degrees]54'E
18. Xora        32[degrees]11'S 29[degrees]00'E
19. Umtata      31[degrees]56'S 29[degrees]11'E
20. Umdumbe     31[degrees]55'S 29[degrees]13'E

                  Oyster Species Present

                 Striostrea    Saccostrea
   Estuary      margaritacea   cuccullata

1. Breede         [check]
2. Duiwenhoks     [check]
3. Goukou         [check]
4. Gourits        [check]
5. Knysna         [check]
6. Keurbooms
7. Kromme         [check]
8. Swartkops      [check]
9. Sundays        [check]
10. Bushmans      [check]
11. Kariega       [check]
12. Kowie         [check]
13. Keiskamma     [check]       [check]
14. Tyolomnga     [check]       [check]
15. Kwelera       [check]       [check]
16. Qora          [check]       [check]
17. Mbashe        [check]       [check]
18. Xora          [check]       [check]
19. Umtata        [check]       [check]
20. Umdumbe       [check]       [check]