Condition, reproductive activity, and gross biochemical composition of the Manila clam, Tapes philippinarum in natural and newly created sandy habitats of the southern coast of Korea.
KEY WORDS: Clam culture, Tapes philippinarum, condition, reproductive cycle, biochemical composition, created habitat
The Manila clam, Tapes philippinarum (Adams & Reeve 1850), is a mollusc species that lives in sand, sandy-silt, or muddy-gravel sediments from the intertidal to subtidal zones a few meters in depth. It is also one of the most widely exploited bivalve resources for human consumption. Because of its high productivity and commercial importance, it is now cultivated worldwide. Numerous studies, particularly for commercial production, have enhanced understanding of the growth, reproduction, and physiological ecology of the clam under different environmental conditions, and thereby enabled rearing outside its normal habitat (Bourne 1982, Beninger & Lucas 1984, Xie & Burnell 1994).
In contrast to a progressive increase in T. philippinarum production in coastal areas where the clam was introduced, a steady trend of decreasing annual production has been observed in indigenous habitats on the Korean coast, from approximately 74,000 tonnes in 1990 to around 38,000 tonnes in 2000 (Ministry of Maritime Affairs and Fisheries Republic of Korea 2005, National Fisheries Research and Development Institute 2004). The major causes of this decline have long been recognized and include habitat loss through reclamation of tidal flats, habitat disturbance by marine pollution, overexploitation of the species, mass mortality (main causes are unknown yet), and predation of mud shrimp on young clams (Chung et al. 1994, Chung et al. 2001). Of these, the loss of clam habitat caused by large-scale reclamation for intertidal flats in Korea has been extensive during the past few decades (Koh 2001). About 40% of a total of 2850 [km.sup.2] of intertidal area has been reclaimed during that period in South Korea. Clam growth is closely related to sediment type, with better growth occurring at low silt content (Goulletquer et al. 1999, Melia et al. 2004), so it may be feasible to create new Manila clam habitats on the muddy intertidal zone and develop clam-farming grounds by adding sand to mudflat areas. Detailed knowledge of the survival, growth, and reproduction of clam populations farmed in newly created sandy habitat is fundamental to improving cultivation of T. philippinarum.
The condition (flesh content) and biochemical composition of bivalves vary in time and space, because they are affected by exogenous factors, such as water temperature and food availability, and endogenous factors, such as energy demands for reproduction (Ansell & Trevallion 1967, Dare & Davies 1975, Newell & Bayne 1980, Navarro et al. 1989, Okumus& Stirling 1998, Kang et al. 2000). A close relationship has also been demonstrated between condition, the accumulation and utilization of nutrients, and the reproductive cycle in various bivalve species (Walne 1970, Gabbott 1975, Bayne 1976, Zandee et al. 1980, Ruiz et al. 1992, Ojea et al. 2004). Accordingly, seasonal variations in condition and biochemical composition of T. philippinarum can give a seasonal indication of its physiological or nutritional state under varying environmental conditions during the gametogenic cycle (Beninger & Lucas 1984, Robert et al. 1993, Marin et al. 2003, Meneghetti et al. 2004).
In this study we evaluated seasonal variations in condition, reproductive activity, and biochemical composition of the Manila clam, T. philippinarum, in natural and newly created sandy habitats.
MATERIALS AND METHODS
Study Site and Field Experiment Design
Jinju Bay is a small (about 25 km long and 13 km wide), shallow bay located on the southern coast of Korea. with a mean water depth of 4 m (Fig. 1). It is well protected by islands to the south and the water is well mixed by strong tidal currents. The tide is semidiurnal with a maximum tidal range of 3.6 m on spring tide. Intertidal sediments in the bay are typically muddy sand, and about 411 ha have been developed for Manila clam culture. However, mud depth in the bay has increased up to 50 cm during the last two decades through sedimentation of suspended silt transported in intermittent freshwater pulses from the artificial dam on the Nam River, north of the bay. This has made the sediment muddier and thus reduced the area of clam habitat.
To assess if clam farming ground could be developed, sand of a total of 900 [m.sup.3] to a depth of 18 cm was added to the fine sediment surface of 5000 [m.sup.2] (50 m x 100 m) of a mudflat site on the west of the bay in March 2000. The sanded flat was tilled by raking nine months after addition of the sand. creating a silty sand habitat. In May 2000, clam juveniles collected from natural beds were seeded onto three sites: the created silty sand habitat, a nonsanded mudflat area, and a natural habitat area. Juvenile clams of about 1500 kg were planted at each site. The period of immersion varied between 1 h (neap tides) and 5 h (spring tides) per tidal cycle at the three sites. Water depth was approximately 2 m at high tide, depending on the neap-spring tidal cycle. Sediment characteristics of the experimental sites, and initial density, shell length, and final density of the clam are presented in Table 1. Initial clam density was 153 [+ or -] 8 (SE). 112 [+ or -] 10, and 112 [+ or -] 10 individuals [m.sup.-2] at the non-sanded mudflat area, created silty sand habitat, and natural habitat, respectively. At the end of experiment, the clam density was 34 [+ or -] 3 (SE) and 23 [+ or -] 1 (SE) individuals [m.sup.-2] at the created silty sand habitat and natural habitat, respectively. Survival rates were over 90% of the initial density at those two sites two months after seeding, but all the clams at the nonsanded mudflat area had disappeared during that period. Therefore, further monthly sampling was only conducted at the created silty sand habitat (the treatment site) and natural habitat (the control site).
[FIGURE 1 OMITTED]
Clam Collection and Biometric Measurement
Clams were collected monthly at the treatment and control sites from May 2000 to October 2001. Clams of similar size were sampled to minimize compositional variations because of age (size) class differences and to assess seasonal variation in clam activities (Fig. 2). About 60 individuals were collected at each site using a manual rake. Specimens were transported to the laboratory and placed overnight in filtered seawater at in situ temperature, to ensure gut evacuation and removal of fecal contents. After rinsing, the weight of each clam was determined and the shell length, width, and height were measured to the nearest 0.05 mm using Vernier calipers. The clams were then dissected and the sex of each determined by a smear. Tissue dry weight and biochemical composition was determined for 15-20 specimens of each sex, and a further 10 specimens of each sex were used for histological analysis. The shell of each clam was dried at 60[degrees]C for approximately 72 h, and the shell weight recorded. Tissues dissected for biochemical analysis were freeze-dried and the tissue dry weight of each individual was determined by subtracting the water content mass from wet tissue mass. The dried tissues were ground to powder with a mortar and pestle and an aliquot was heated at 450[degrees]C for 24 h to determine ash weight. The remaining dry tissue was stored at -30[degrees]C in a refrigerator for later biochemical analysis.
On each sampling occasion at each site the temperature and salinity of surface and bottom waters were measured at high tide using a CTD meter (Seabird Electronics, Inc.). For bottom waters, 5 L was pumped from 1 m above the bottom, screened through a 180 [micro]m Nitex mesh to eliminate zooplankton and large particles, and collected in acid-washed plastic bottles. Water samples were kept on ice in the dark until filtered on pre-combusted Whatman GF/F glass-fiber filters (47 mm, 0.7 [micro]m pore size). Filters for chlorophyll a analysis were stored at -80[degrees]C, and filters for suspended particulate matter were dried at 60[degrees]C overnight and then placed in a dessicator.
Chlorophyll a concentration was determined from acetone extracts using a fluorometric method according to Holm-Hansen et al. (1965), using a fluorometer (Turner Designs Model 10 AU 005). Total suspended particulate matter (SPM) was determined from filter weights before and after drying, following filtration of a known volume of water.
Condition and Reproductive Cycle
The condition index was calculated from the dry weight of tissue and shell according to the formula: Condition = [tissue dry weight (mg)/dry shell weight (mg)] x 100 (Lucas and Beninger 1985).
For 10 specimens of each sex, the gonads were fixed in Bouin solution, embedded in paraffin, sectioned at 5 [micro]m, and stained with iron hematoxylin-eosin (Humason 1962). The gametogenic stage was classified (Chung et al. 1994, Sbrenna & Campioni 1994), and scored on a 0-4 scale according to Mann's (1979a) scheme: stage 0 = inactive, stage 1 = early active, stage 2 = late active, stage 3 = ripe, stage 4 = spawning-spent). The arithmetic means of the individual scores of whole specimens were recorded as the gonadal maturity index (GMI) for each sampling occasion (see details in Meneghetti et al. 2004).
Powdered sample (5-10 mg) was used for biochemical analysis. Protein content was determined using the method of Lowry et al. (1951) after alkaline hydrolysis with 0.5 N NaOH at 30[degrees]C for 24 h. Carbohydrate and glycogen were extracted in 15% trichloroacetic acid and determined as glucose following the phenol-sulfuric acid method (Dubois et al. 1956). Glycogen was quantified after precipitation with 100% ethanol. Extraction of total lipid was performed in a mixture of chloroform and methanol (Bligh & Dyer 1959) and lipid content determined following the method of Marsh and Weinstein (1966).
[FIGURE 2 OMITTED]
To evaluate the physiological state of T. philippinarum independent of growth, absolute values for tissue dry weight were standardized to that of a clam (standard animal) of 31.5 mm in shell length (the mean of all specimens analyzed) and compared for each sampling occasion. Allometric equations of tissue dry weight against shell length at each sampling were determined by linear regression analysis following logarithmic transformation (base 10). The same analysis was used to relate the weight of biochemical constituents and ash to the tissue dry weight. All regressions were statistically significant (P < 0.01). Dry weights of biochemical constituents were then computed for a given shell length by substituting appropriate values of tissue weight in the regression equations referred to above (see details in Navarro et al. 1989). The results of biochemical analysis were then expressed in mg per standard animal.
The clam tissue [[delta].sup.13]C and [[delta].sup.15]N were measured for at least four individuals, sampled bimonthly at each site as described above. Marine particulate organic matter (POM) and riverine POM were collected bimonthly at the bay mouth and downstream of the Nam River. For sampling POM at each site, about 20 L of water was prefiltered in situ with a 250-[micro]m screen to remove large particles and transported to the laboratory as soon as possible. The particulates in this prefiltered water were concentrated onto a precombusted Whatman GF/F filter in the laboratory, treated with 2-3 drops of 10% HCl, and then rinsed with distilled water. For each POM sampling, microphytobenthos was collected by scraping the visible mat of benthic diatoms from the sediment surface. This was extracted and prepared for isotope analysis according to Couch's procedure (1989), as described by Riera and Richard (1996). These pretreated samples were kept frozen until analysis, freeze-dried at -70[degrees]C, and then loaded into tin capsules prior to isotope analysis.
Values for [[delta].sup.13]C and [[delta].sup.15]N were obtained using a CHN elemental analyzer (EuroVector 3000 Series, Italy) combined with a continuous-flow isotope ratio mass spectrometer (GV Instruments, IsoPrime, UK). All isotope ratios were expressed as the relative per mill ([per thousand]) difference between the sample and conventional standard reference materials (PeeDee Belemnite carbonate and [N.sub.2] in air) as follows: [delta]X = [([R.sub.sample]/[R.sub.standard]) - 1] x [10.sup.3], where X is [sup.13]C or [sup.15]N, and R is the corresponding ratio of [sup.13]C:[sup.12]C or [sup.15]N:[sup.14]N. A laboratory internal standard (peptone, Merke) was run every sixth sample. Measurement precision based on the standard deviation of 20 replicates of the internal standard was within 0.2[per thousand] for both isotope pairs. Two replicates were analyzed from each sample of ground clam tissue.
Commercially available software was used to analyze the experimental results (SPSS package, Chicago, IL). Data were tested for normality using the Shapiro-Wilk procedure. Leven test was used to check the homogeneity of variance among data. A paired sample t-test for paired comparisons was used to test the significance of variations in environmental parameters between sites and to determine differences in condition index, tissue dry weight, and biochemical components between sexes or sites during the study period. The arcsine transformation was applied to percentage data, such as condition index, prior to analysis. Analysis of variance (ANOVA, two-way) with Tukey multiple comparison test was performed to determine differences in [[delta].sup.13]C and [[delta].sup.15]N values of the Manila clam tissues between sites and among sampling months.
Temperature, Salinity, and Potential Food
Monthly mean water temperature showed a seasonal cycle typical of temperate zones, with maximum temperatures (22.6[degrees]C to 25.1[degrees]C) in July-September and minimum temperatures (7.4[degrees]C to 8.0[degrees]C) in January-February (Fig. 3). Salinity varied inversely with a summer minimum of less than 30 PSU in 2000 and 31 PSU in 2001, according to the monsoonal climate and constant values at around 33 PSU for the major part of the year. No differences in temperature and salinity were found between sites (paired t-test, P = 0.798 and 0.067).
Monthly mean SPM concentration fluctuated between 6.9 and 19.0 mg [L.sup.-1] (Fig. 4A). Although relatively high concentrations of more than 16 mg [L.sup.-1] were recorded during May to July 2001, concentrations were highly variable. Although chlorophyll a concentration peaked in summer in 2000, coincident with the period of salinity minima, its seasonal trend was not clear (Fig. 4B). Rather, chlorophyll a concentration was also characterized by irregular peaks as shown in variations of SPM concentration. Both SPM and chlorophyll a values did not differ significantly between sites (P = 0.422 and 0.559).
[FIGURE 3 OMITTED]
Condition and Reproductive Cycle
Because of the possibility of sexual differentiation in biochemical composition, as shown in the clam Ruditapes decussates (L.) (Perez Camacho et al. 2003), all analyses in this study were carried out separately for each sex. Seasonal variations in condition index were very similar between control and treatment sites, and the data showed a clear seasonal pattern (Fig. 5) that was very similar between sexes. The maximum values in condition index occurred in May 2000 and were followed by a progressive decline throughout the summer-autumn period to minimum levels in October. After November a subsequent increase was observed, which peaked in March to April 2001. This spring peak was followed by a summer decline. No statistical difference in condition index was found between females and males of each population (P = 0.583 and 0.113 for control and treatment sites, respectively) or between sites (P = 0.466 and 0.791 for female and male, respectively).
Values for GMI also exhibited clear seasonality that was similar between control and treatment sites (Fig. 6). This was attributed to sexual maturation of the clams. In both years, the development of gonadal tissue was initiated in both sexes in January and peaked during May to July, when the condition index values started to decline. Gametogenic stage 4 can be used as the criterion for the beginning and end of spawning. The appearance and disappearance of stage 4 coincided with the period of condition decline, indicating that spawning began in May and continued until October. Spawning was followed by a sexual resting period in November to December.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Tissue Dry Weight
The tissue dry weight of a standard animal was calculated for each sex in the control and treatment groups (Fig. 7). There was a remarkable seasonal variation in tissue dry weight in both groups, the patterns being similar to those of the condition index. The dry weight of soft tissues decreased steadily from maximum levels in May 2000 to minimum levels in October, coincident with the spawning period, and increased progressively to spring 2001, when values peaked. This was followed by a gradual decrease to the end of the experiment in October 2001. Differences in tissue dry weight of standard animals were not statistically significant between females and males of each population (P = 0.150 and 0.727 for control and treatment sites) or between sites (P = 0.145 and 0.877 for female and male).
Gross Biochemical Composition
Seasonal variations in gross biochemical composition of standard animals showed similar patterns between sexes of each population or sites, with clear seasonal trends (Fig. 8). Seasonal patterns in absolute values for all biochemical components (protein, carbohydrate, and lipid) were very similar and followed those of tissue dry weight of standard animals and the condition index for each population.
Protein peaked at the beginning of the study in May 2000 and in April 2001. Sharp declines occurred in June 2000 and from August to October, when minimum values were detected. Loss in protein content correlated with spawning. Values then increased gradually to peak in March to April 2001, followed by a sudden decline in May and July to October. No statistical difference was found in the protein value of standard animals between females and males of each population (P = 0.148 and 0.811 for control and treatment sites) or between sites (P = 0.121 and 0.182 for female and male).
A similar seasonal pattern for carbohydrate and glycogen indicated that carbohydrate content of the clams depends on glycogen mobilization. Similar to protein, a carbohydrate peak occurred in May 2000 and was followed by a sudden decrease from June until October to November. Carbohydrate (glycogen) reserves were quickly restored during the winter to spring and decreased again through the summer to a minimum in October 2001. A paired t-test showed no significant difference in carbohydrate and glycogen values between females and males of each population (P = 0.693 and 0.820 for carbohydrate for control and treatment sites); 0.123 and 0.135 for glycogen) or between sites (P = 0.827 and 0.727 for carbohydrate for females and males; 0.199 and 0.142 for glycogen, respectively).
Lipid values were lowest in October to November and highest in spring, displaying a similar seasonal pattern to protein and carbohydrate. There was no difference in the lipid value of standard female and male clams between each population (P = 0.094 and 0.331 for control and treatment sites) or between sites (P = 0.389 and 0.341 for female and male).
Although peak values of ash content were observed in spring, its seasonal pattern was less pronounced than other biochemical variables. No significant difference in ash value of standard animals was found between females and males of each population (P = 0.141 and 0.191 for control and treatment sites) or between sites (P = 0.210 and 0.090, for females and males).
Stable Isotope Ratios for Clams and Potential Food Sources
Values for [[delta].sup.13]C in muscle tissue of clams averaged -17.4 [+ or -] 0.6[per thousand] (SD, n = 26) and -17.6 [+ or -] 0.6[per thousand] (n = 25) at the control and treatment sites, respectively (Table 2). There were no significant differences in [[delta].sup.13]C values between sites and among the sampling months (two-way ANOVA, [F.sub.(1, 39)] = 0.059, P = 0.809 for site; [F.sub.5, 39] = 2.318, P = 0.062 for month). Values for [[delta].sup.15]N averaged 10.8 [+ or -] 0.8[per thousand] (n = 23) and 10.5 [+ or -] 0.8[per thousand] (n = 23) at the control and treatment sites, respectively (Table 2), with no significant difference between sites (two-way ANOVA, [F.sub.5, 34] = 20.186, P < 0.001). However, the [[delta].sup.15]N values showed significant seasonal differences, being about 2[per thousand] more depleted in June 2001 than in August, October, and December 2000 for both sites ([F.sub.5, 39] = 2.318, P = 0.062; Tukey HSD test, P < 0.05). Although significant seasonal changes in both [[delta].sup.13]C and [[delta].sup.15]N values for microphytobenthos, marine POM, and riverine POM were detected, the stable isotope compositions of these three potential food sources were clearly distinguished: the average values (n = 12) were, respectively, -14.5 [+ or -] 1.6[per thousand], 21.1 [+ or -] 1.1[per thousand], and -27.8 [+ or -] 2.7[per thousand] for [[delta].sup.13]C, and 8.9 [+ or -] 0.9[per thousand], 5.7 [+ or -] 1.3[per thousand], and 9.0 [+ or -] 0.8[per thousand] for [[delta].sup.15]N (Table 2; Fig. 9).
According to a dual plot of [[delta].sup.13]C and [[delta].sup.15]N deviations measured for the clams and their potential food sources (Fig. 9), the dietary components of the clams appeared to be primarily composed of microphytobenthos and marine POM throughout the year. Figure 10 presents the relative biomass contributions of the three dietary sources, calculated from a linear mixing model based on mass balance equations using [[delta].sup.13]C and [[delta].sup.15]N values (the IsoSource model, Phillips & Gregg 2003). The results confirmed that a major part of clam biomass for both sites was derived from microphytobenthos and marine POM. Microphytobenthos contribution tended to be relatively constant from August 2000 to April 2001, and then sharply decreased to June. Conversely, the contribution of marine POM progressively increased from October 2000 to June 2001. The dietary contribution of riverine POM ranged between 11% and 17% in August and October 2000, but it was negligible throughout the rest of the year.
[FIGURE 6 OMITTED]
The condition index of marine bivalves summarizes the physiological activity of the animals under given environmental conditions, and its fluctuation has important implications for cultivation and harvesting strategies for bivalves (Lucas & Beninger 1985, Okumus& Stirling 1998). The primary objective in this study was to compare the condition of Manila clams (T. philippinarum) reared in natural and artificially created habitat in relation to gametogenic processes, and accumulation and mobilization of reserve materials, as well as to evaluate habitat functions of the created farming grounds for this species. The condition index had a seasonal cycle at both the treatment and the control site, with minimum values in autumn (October to November) rising to peak levels in early spring (May 2000 and March to April 2001) (Fig. 5). No difference was evident in the index level between clams from the natural and created habitat. As the condition of bivalves reflects changes of physiological status dependent on environmental factors including food supply and temperature, this result indicates that clam stocks from created silty sand habitat can have biological traits similar to those from natural clam habitat and provide a good quality product for market.
In this study, marine sand transported by a large barge was sprinkled on the muddy sediments during the high slack water period and then flattened during the ebb tide. Adding sand made the muddy substrate coarser (Table 1). Monthly monitoring of the sediment characteristics showed that chemical oxygen demand (7.53-11.80 mg [g.sup.-1]), hydrogen sulfide (0.00-0.08 mg [g.sup.-1]), and organic content (1.0-7.7%) of coarser sediments in the created habitats were much lower than those (9.27-32.58 mg [g.sup.-1], 0.00-0.10 mg [g.sup.-l], and 7.0-8.1%, respectively) in muddy sediments (data not shown in Results). Clam-farming ground exposed to environmental conditions similar to those of natural clam habitat was created by adding sand to the muddy substrate, and none of the hydrological and sedimentary parameters evaluated showed significant differences between the natural and created habitats (Figs. 3 and 4, Table 1). Although most differences were caused by reproductive and physiological characteristics of clams rather than regional differences in environmental conditions, environmentally stressful conditions can result in differences in physiological status of marine bivalves in time and space (Navarro et al. 1989, Robert et al. 1993, Kang et al. 2000, Marin et al. 2003). Accordingly, the comparable environmental conditions between the natural and created habitat were probably responsible for the similarity in timing and level of the maximum and minimum values for the condition index of clams in this study.
[FIGURE 7 OMITTED]
Spring maxima in condition found during this investigation have also been observed in European populations (Beninger & Lucas 1984, Marin et al. 2003). Biometric data revealed that the maximum condition value was considerably greater in both stocks in the spring of 2000 than in 2001. Similar results were obtained by Beninger and Lucas (1984) in T. philippinarum from the north Atlantic coast of France. This may be explained by the diminishing growth rate with size (age) in marine bivalves (Ohba 1959, Beninger & Lucas 1984, and references therein). The spring maxima in condition were followed by a progressive decline during summer and autumn caused by spawning, after which clam condition recovered rapidly in parallel to accumulation of energy reserves in late autumn to winter 2000. The timing of condition minima recorded during this study appeared to be somewhat different from those observed in European clam populations. In Brittany (France), Beninger and Lucas (1984) observed only a partial condition recovery in spring because of physiological stress during the autumn and winter. In the Lagoon of Venice, Marin et al. (2003) also found condition recovery from February throughout spring to summer, because of the decrease in water temperature and food availability in autumn to winter. In other words, minimum values in condition index were recorded during winter in European waters. These differences in timing of condition minima and recovery suggest that the clams at the study sites had a positive energy balance during late autumn and winter, as indicated by the accumulation of energy reserves after spawning.
Strong positive correlations between condition index and GMI values (control: r = 0.752 and 0.780 for female and male, P < 0.001 for both; treatment: r = 0.748 and 0.724, P < 0.001) indicated a close relationship between the gametogenic cycle and condition. There was no apparent difference in stages of gonadal maturation between sites (Fig. 6). After a brief inactive period in late autumn (November to December), gametogenesis was initiated in January. Ripe stage was reached in spring and was followed by spawning, which began in May and lasted throughout summer until October. Mean water temperature was 7.6[degrees]C in January 2001 and 14.3[degrees]C in May, comparable to minimum temperatures of 8[degrees]C for gonadal activity and 14[degrees]C for spawning of T. philippinarum (Mann 1979b). The reproductive cycle of T. philippinarum observed during this investigation is very similar to that of Mediterranean populations (Sbrenna & Campioni 1994, Meneghetti et al. 2004) and a Japanese population (Ohba 1959). However, the onset of gametogenesis and gamete ripening in northern European populations (Beninger & Lucas 1984, Xie & Burnell 1994) and Vostok Bay populations of the northwestern part of the East/Japan Sea (Ponurovsky & Yakovlev 1992) is delayed compared with populations in warmer climates. Xie and Burnell (1994) speculated that this time lag could be explained by a time to temperature effect.
Continuous or multiple spawning during summer has been observed worldwide (Ponurovsky & Yakovlev 1992, Sbrenna & Campioni 1994, and references therein). In the present investigation, it is difficult to determine the precise timing and intensity of spawning from a monthly sampling strategy. The seasonal cycles of gametogenesis and the amplitude of seasonal fluctuations of standard animal tissue dry weight was quite similar between the natural and created habitat. The loss of tissue weight observed during the spring-summer period appeared to coincide with spawning time as seen in the seasonal variations in the sexual maturation process and the GMI values (Figs. 6 and 7). The rapid decrease in standard animal tissue dry weight during the spring to summer period suggests two spawning events. The spawning pattern of the Manila clam is known to be highly variable, with one spawning period reported in Washington State and Ireland (Holland & Chew 1974, Xie & Burnell 1994, respectively); two spawning peaks in Japan (Tanaka 1954, Ohba 1959), France (Beninger & Lucas 1984) and Italy (Sbrenna & Campioni 1994, Meneghetti et al. 2004); and three spawning events in southwest Spain (Sarasquete et al. 1990). The timing and intensity of gonadal development and spawning is known to vary locally, seasonally, and interannually (Beninger & Lucas 1984, Navarro et al. 1989, Sbrenna & Campioni 1994, Kang et al. 2000, Meneghetti et al. 2004). These studies demonstrated that temperature is not the only factor acting on gonad development and the beginning of spawning, and that local or temporal variations in gonadal maturation of bivalves are closely related to variations in nutritional conditions (that is, food availability). Therefore, the gametogenic cycle observed during this investigation appears to be explained by a seasonal cycle in accumulation and mobilization of reserve materials in clams in relation to variability in trophic conditions.
[FIGURE 8 OMITTED]
Seasonal patterns and levels of tissue dry weight and biochemical components in clams were quite similar in the natural and created habitat (Figs. 7 and 8), indicating that clams reared in created habitats can have the same physiological characteristics as those in natural habitats. Seasonal variations in the absolute values of biochemical components paralleled those of tissue dry weights in clams at both sites. Seasonal cycles in tissue dry weight of standard animals seemed to reflect the reproductive cycle, with weight being maximal prior to the spring to summer spawning, and then decreasing simultaneously with spawning. Protein constitutes the major organic component of gametes in T. philippinarum (Beninger & Lucas 1984). Somatic proteins are also mobilized during gametogenesis, acting as the predominant respiratory substrate during this period (Mann & Glomb 1978, Adachi 1979; see also Beninger & Lucas 1984, Marin et al. 2003). As expected from these studies, the absolute values for protein in clams in this study peaked during the spring-summer spawning period (Fig. 8). A sharp decline in May to June 2000 and April to May 2001 indicated the first spawning during these periods. After this first spawning, there was a rapid recovery in protein content and then a slow decrease until October, indicating the second spawning event.
[FIGURE 9 OMITTED]
Seasonal variation in absolute values for lipid was very similar to those of protein. It is commonly accepted that glycogen reserves are converted into lipids, which are a major component of bivalve oocytes (Holland 1978, Gabbott 1983, Beninger & Lucas 1984, Marin et al. 2003). Accordingly, the maximum values detected prior to spawning, and the minimum after, may be explained by the above hypothesis. As observed in the seasonal variation in protein, there was a rapid recovery in lipid content after the first spawning. Somatic proteins and lipids also serve as energy reserves in T. philippinarum under conditions of nutritional stress and energy imbalance (Beninger & Lucas 1984, Marin et al. 2003). After the spawning period in this study, tissue dry weight began to increase without a period of decrease during inactive gonadal phase (no oocytes) in winter, in contrast to European populations (Beninger & Lucas 1984, Marin et al. 2003). Because the absolute values of all biochemical components also increased during this period, it is difficult from this study to confirm the role of proteins and lipids as maintenance energy reserves.
Seasonal patterns of carbohydrate in clams from the sampling sites could be explained by glycogen patterns. Carbohydrate and glycogen peaks in May 2000 were followed by a sudden decrease as a consequence of spawning. In contrast to protein and lipid, glycogen recovery was not detected after the first spawning in both 2000 and 2001. Glycogen has long been considered to be the main energy reserve for both the formation of gametes and the maintenance of adult bivalves during periods of nutritional stress (see Beninger & Lucas 1984). The results presented here support this conclusion and also the conversion of glycogen into lipids, as previously discussed. The absolute values for glycogen were lowest at the end of spawning and then quickly recovered throughout winter to spring. In general, the recovery and accumulation of glycogen in bivalves is related to good conditions of food availability (Ansell & Trevallion 1967, Newell & Bayne 1980, Navarro et al. 1989, Okumus& Stirling 1998, Kang et al. 2000). Beninger and Lucas (1984) showed that a failure to constitute energy reserves in autumn after spawning, and the exhaustion of reserves caused by energy imbalances in winter, resulted in a delay in the recovery of condition and the maturation of gonads in Brittany (France) populations of T. philippinarum. On the other hand, Sbrenna and Campioni (1994), and Meneghetti et al. (2004) suggested that high accumulation of reserve materials, because of high food availability from late winter to early spring, leads to gamete development in winter and spawning in May in lagoon systems of Italy. Thus, the recovery of glycogen reserves throughout winter--spring in this study indicates that the Manila clams had sufficient food supply to enable resumption of gonadal maturation and spawning in May.
[FIGURE 10 OMITTED]
Comparison of isotopic signatures of T. philippinarum tissues with those of potential food resources confirmed the contribution of microphytobenthos to the clam diet in both the natural and the created habitat (Fig. 10). The POM and chlorophyll a data did not show any seasonal trend, indicating no apparent seasonality in the trophic conditions of ambient waters. Irregular peaks in SPM and chlorophyll a values, as well as the isotopic data. suggest that the nutritional conditions of waters were strongly influenced by resuspension caused by tide and wind. Therefore, despite an important role of phytoplankton in the clam diet throughout the year, food availability to meet energy requirements in this study area might be maintained by resuspension of microphytobenthos, even during the winter to spring period. The high availability of food because of the summer phytoplankton blooms may have provided energy requirements for the rapid recovery of reserve materials after the first spawning as well as for the second spawning in summer. The isotopic signatures of clam tissues showed that the dietary contribution of organic matter from marine POM (largely phytoplankton) increased markedly in late spring to summer 2001. Phytoplankton blooms caused by heavy rain and nutrient input, which are concentrated during late spring to early summer, are well-known phenomena in coastal regions with a monsoonal climate, as occurs on the Korean peninsula (Kang et al. 2000, Park et al. 2001). Although limited to late summer, a dietary contribution from riverine POM was detected. This is consistent with the conclusion of Kasai et al. (2004) that Manila clams select marine POM (phytoplankton and microphytobenthos) from the organic matter available in their habitat, and the contribution of terrestrial material increases temporarily during rainfall. Finally, our biochemical and isotopic data indicated that food availability in the study area was mostly dependent on resuspension of microphytobenthos, along with seasonal dynamics of phytoplankton.
In summary, clams reared in newly created or natural habitats had similar patterns and levels in condition and tissue dry weight. The similar biochemical compositions and reproductive cycles for clams in the two habitats are likely to be a consequence of similarities in environmental conditions, and thereby food availability. The results show that the newly created sandy habitats may provide habitat functions that enable Manila clams to have biological cycles similar to those in natural habitats. These results confirm the conclusion of Melia et al. (2004) that rearing sites with high sand content are preferable in terms of growth and maximum attainable size of Manila clams. This initial experiment primarily tested whether Manila clams can adapt to created sandy habitats. For commercially sustainable exploitation and better yield of T. philippinarum in the future, more detailed studies of the effect of stock density and deposition rate of silt modifying the substrate are needed.
The authors thank H. J. Park, J. H. Kwak, and E. J. Jang for their technical assistance and the anonymous referees for their reviewing and critical comments on the manuscript. This work was supported for two years by Pusan National University Research Grant.
Adachi, K. 1979. Seasonal changes of the protein level in the adductor muscle of the clam, Tapes philippinarum (Adams and Reeve) with special reference to the reproductive seasons. Comp. Biochem. Physiol. 64:85-89.
Ansell, A. D. & A. Trevallion. 1967. Studies on Tellina tenuis de Costa: I. Seasonal growth and biochemical cycle. J. Exp. Mar. Biol. Ecol. 1:220-235.
Bayne, B. L. 1976. Aspects of reproduction in bivalve mollusks. In: M. L. Wiley, editor. Estuarine processes. New York: Academic Press. pp. 432-448.
Beninger, P. G. & A. Lucas. 1984. Seasonal variations in condition, reproductive activity, and gross biochemical composition of two species of adult clam reared in a common habitat: Tapes decussates (L.) (Jeffeys) and Tapes philippinarum (Adams & Reeve). J. Exp. Mar. Biol. Ecol. 79:19-37.
Bligh, E. G. & W. F. Dyer. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911-917.
Bourne, N. 1982. Distribution, reproduction, and growth of Manila clam, Tapes philippinarum (Adams & Reeves), in British Columbia. J. Shellfish Res. 2:47-54.
Chung, E. Y., D. K. Ryou & J. H. Lee. 1994. Gonadal development, age and growth of the shortnecked clam, Ruditapes philippinarum (Pelecypoda: Veneridae), on the coast of Kimje, Korea. Korean J. Malucol 10:38-54.
Chung, E. Y., S. B. Hur, Y. B. Hur & J. S. Lee. 2001. Gonadal maturation and artificial spawning of the Manila clam, Ruditapes philippinarum (Pelecypoda: Veneridae), in Komso Bay, Korea. J. Fish. Sci. Tech. 4:208-218.
Couch, C. A. 1989. Carbon and nitrogen stable isotopes of meiobenthos and their food resources. Estuar. Coast. Shelf Sci. 28:433-441.
Dare, P. J. & G. Davies. 1975. Experimental suspended culture of mussels, Mytilus edulis (L), in Wales using spat transplanted from a distant settlement ground. Aquaculture 6:257-274.
DeNiro, N. J. & S. Epstein. 1979. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42:495-506.
DeNiro, N. J. & S. Epstein. 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta 45:341-351.
Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebecs & F. Smith. 1956. Colorimetric method for the determination of sugars and related substances. Anal. Chem. 28:350-356.
Gabbott, P. A. 1975. Storage cycles in marine bivalve mollusks: an hypothesis concerning the relation between glycogen and gametogenesis. In: H. Barnes, editor. Proceedings of the Ninth European Marine Biology Symposium. Aberdeen, Scotland, UK: Aberdeen University Press. pp. 191-211.
Gabbott, P. A. 1983. Development and seasonal metabolic activities in marine mollusks. In: A. S. M. Saleuddin & K. M. Wilbur, editors. The Mollusca. Vol. 2. New York: Academic Press. pp. 165-217.
Goulletquer, P., R. Robert & G. Trut. 1999. Manila clam Tapes philippinarum culture: sediment-clam interactions. Aquat. Living Resour. 12:45-56.
Holland, D. A. & K. K. Chew. 1974. Reproductive cycle of the Manila clam (Venerupis japonica), from Hood Canal, Washington. Proc. Nat. Shellfish Ass. 64:53-58.
Holland, D. L. 1978. Lipid reserves and energy metabolism in the larvae of benthic marine invertebrates. In: D. C. Malins & J. R. Sargent, editors. Biochemical and Biophysical Perspectives in Marine Biology. London: Academic Press. pp. 85-123.
Holm-Hansen, O., C. J. Lorenzen, R. W. Holms & J. D. H. Strickland. 1965. Fluorometric determination of chlorophyll. J. Cons. Perm. Int. Explor. Mer. 30:3-15.
Humason, G. L. 1962. Animal tissue techniques. San Francisco: W H Freeman & Co.
Kang, C. K., M. S. Park, P. Y. Lee, W. J. Choi & W. C. Lee. 2000. Seasonal variations in condition, reproductive activity, and biochemical composition of the Pacific oyster, Crassostera gigas (Thunberg), in suspended culture in two coastal bays of Korea. J. Shellfish Res. 19:771-778.
Kasai, A., H. Horie & W. Sakamoto. 2004. Selection of food sources by Ruditapes philippinarum and Mactra veneriformis (Bivalva: Mollusca) determined from stable isotope analysis. Fish. Sci. 70: 11-20.
Koh, C. H. 2001. The Korean tidal flat: environment, biology and human. Seoul, Korea: Seoul National University Press. 1073 pp.
Lowry, O. M., N. I. Roseborough, A. L. Farrand & R. J. Randall. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:263-275.
Lucas, A. & P. G. Beninger. 1985. The use of physiological condition indices in marine bivalve aquaculture. Aquaculture 44:187-200.
Mann, R. & S. J. Glomb. 1978. The effect of temperature on growth and ammonia excretion of the Manila clam Tapes japonica. Est. Coast. Mar. Sci. 6:335-339.
Mann, R. 1979a. Some biochemical and physiological aspects of growth and gametogenesis in Crassostrea gigas and Ostrea eclulis grown at sustained elevated temperature. J. Mar. Biol. Ass. UK. 59:95-100.
Mann, R. 1979b. The effect of temperature on growth, physiology, and gametogenesis in the Manila clam Tapes philippinarum (Adams & Reeve, 1850). J. Exp. Mar. Biol. Ecol. 37:121-133.
Marin, M. G., V. Moschino, M. Deppieri & L. Lucchetta. 2003. Variations in gross biochemical composition, energy value and condition index of T. philippinarum from the Lagoon of Venice. Aquaculture 219:859-871.
Marsh, J. B. & D. B. Weinstein. 1966. Simple charring method for determination of lipid. J. Lipid Res. 7:57-576.
Melia, P., G. A. De Leo & M. Gatto. 2004. Density and temperature-dependence of vital rates in the Manila clam Tapes philippinarum: a stochastic demographic model. Mar. Ecol. Prog. Ser. 272:153-164.
Meneghetti, F., V. Moschino & L. Da Ros. 2004. Gametogenic cycle and variations in oocyte size of Tapes philippinarum from the Lagoon of Venice. Aquaculture 240:473-488.
Minagawa, M. & E. Wada. 1984. Stepwise enrichment of [sup.15]N along food chains: further evidence and the relation between [delta][sup.15]N and animal age. Geoehim. Cosmochim. Acta 48:1135-1140.
Ministry of Maritime Affairs and Fisheries Republic of Korea. 2005. 2005 Statistical year book of maritime affairs and fisheries, http:// fs.fips.go.kr/index.jsp.
National Fisheries Research and Development Institute. 2004. Korea US Aquaculture. http://www.lib.noaa.gov/korea/main_species/ manila.htm.
Navarro, E., J. I. P. Iglesias & A. Larranaga. 1989. Interannual variation in the reproductive cycle and biochemical composition of the cockle Cerastaderma edule from Mundaca Estuary (Biscay, North Spain). Mar. Biol. 101:503-511.
Newell, R. I. & B. L. Bayne. 1980. Seasonal changes in the physiology, reproductive condition and carbohydrate content of the cockle Cardium (= Cerastoderma) edule (Bivalve: Cardiidae). Mar. Biol. 56:11-19.
Ohba, S. 1959. Ecological studies in the natural population of a clam Tapes japonica, with special reference to seasonal variations in the size and structure of the population and to individual growth. Biol. J. Okayama Univ. 5:13-47.
Ojea, J., A. J. Pazos, D. Martinez, S. Novoa, J. L. Sanchez & M. Abad. 2004. Seasonal variation in weight and biochemical composition of the tissues of Ruditapes decussatus in relation to the gametogenic cycle. Aquaculture 238:451-468.
Okumus, I. & H. P. Stirling. 1998. Seasonal variations in the meat weight, condition index and biochemical composition of mussels (Mytilus edulis L.) in suspended culture in two Scottish sea lochs. Aquaculture 159:249-261.
Park, M. S., C. K. Kang & P. Y. Lee. 2001. Reproductive cycle and biochemical composition of the ark shell Scapharca broughtonii (Schrenck) in a southern coastal bay of Korea. J. Shellfish Res. 20:177-184.
Perez Camacho, A., M. Delgado, M. J. Fernandez-Reiriz & U. Labarta. 2003. Energy balance, gonad development and biochemical composition in the clam Ruditales decussatus. Mar. Ecol. Prog. Ser. 258:133-145.
Phillips, D. L. & J. W. Gregg. 2003. Source partitioning using stable isotopes: coping with too many sources. Oecologia 136:261-269.
Ponurovsky, S. K. & Yu. M. Yakovlev. 1992. The reproductive biology of the Japanese littleneck, Tapes philippinarum (A. Adams & Reeves 1850) (Bivalvia: Veneridae). J. Shellfish Res. 11:265-277.
Riera, P. & P. Richard. 1996. Isotopic determination of food sources of Crassostrea gigas along atrophic gradient in the estuarine bay of Marennes-Oleron. Estuar. Coast. Shelf Sci. 42:347-360.
Robert, R., G. Trut & J. L. Laborde. 1993. Growth. reproduction and gross biochemical composition of the Manila clam Ruditapes philippinarum in the Bay of Arcachon, France. Mar. Biol. 116:291-299.
Ruiz, C., D. Martinez, G. Mosquera, M. Abad & J. L. Sanchez. 1992. Seasonal variations in condition, reproductive activity and biochemical composition of the flat oyster, Ostrea edulis, from San Cibran (Galicia, Spain). Mar. Biol. 112:67-74.
Sarasquete, M. C., S. Gimeno & M. L. Gonzalez de Canales. 1990. Cycle reproducteur de la palourde Ruditapes philippinarum (Adams & Reeves 1850) de la cote sud ouest atlantique (Espagne). Rev. hit. Oceanogr. Med. 97-98:90-99.
Sbrenna, G. & D. Campioni. 1994. Gametogenic and spawning patterns of Manila clams Tapes philippinarum (Bivalvia: Veneroida) in two lagoons of the River Po Delta, Italy. J. Shellfish Res. 13:37-46.
Tanaka, Y. 1954. Spawning season of important bivalves in Ariake Bay III Venerupis semidecussata (Reeve). Bull. Jap. Soc. Sci. Fish. 19:1165-1167.
Walne, P. R. 1970. The seasonal variation of meat and glycogen content of seven populations of oysters Ostrea edulis L. and a review of literature. Fishery Invest. Lond. Ser. 2:35.
Xie, Q. & G. M. Burnell. 1994. A comparative study of the gametogenic cycles of the clams Tapes philippinarum (A. Adams & Reeve 1850) and Tapes decussates (Linnaeus) on the south coast of Ireland. J. Shellfish Res. 13:467472.
Zandee, D. I., J. H. Kluytmans, W. Zurburg & H. Pieters. 1980. Seasonal variations in biochemical composition of Mytilus edulis with reference to energy metabolism and gametogenesis. Neth. J. Sea Res. 14:1-29.
CHANG-KEUN KANG, (1) * YANG SOON KANG, (2) EUN JUNG CHOY, (1) DONG-SUN KIM, (3) BONG-TAEK SHIM (4) AND PIL-YONG LEE (2)
(1) Department of Biology, Pusan National University, Busan 609-735, Republic of Korea; (2) Department of Oceanography, National Fisheries Research and Development Institute, Busan 619-900, Republic of Korea; (3) Research Center for Ocean Industrial Development (RCOID), Pukyong National University, 608-737 Busan, Republic of Korea; (4) Goseong Maritime and Fisheries Office, Masan Regional Maritime Affairs & Fisheries Office. 638-805 Gyeongnam, Republic of Korea
* Corresponding author. E-mail: email@example.com
TABLE 1. Sediment characteristics of the experimental sites, and initial density, shell length, and final density of the clam. Data are from May 6 2000, when the sites were seeded with clams, except for final density (October 30 2001). Data are means (ranges in parentheses). Organic % Mud Median Grain Content Site <63 [micro]m Size ([micro]m) (% Ignition loss) Nonsanded mudflat 83 23 7.5 (7.0-8.2) Created habitat 17 193 2.5 (1.0-7.7) Natural habitat 20 185 2.6 (1.0-8.0) Initial (seed) Density Final Density (individuals Initial Shell (Individuals Site [m.sup.-2]) Length (mm) [m.sup.-2]) Nonsanded mudflat 153 (138-164) 25.3 (18.0-33.9) 0 Created habitat 112 (100-131) 24.8 (17.6-41.2) 34 (22-39) Natural habitat 100 (92-123) 24.5 (20.0-38.1) 23 (20-28) TABLE 2. [[delta].sup.13]C and [[delta].sup.15]N values (mean [+ or -] SD) for microphytobenthos, marine POM (particulate organic matter), riverine POM, and clams collected in Jinju Bay. Figures in parentheses are the number of analyzed samples. August Sample 2000 [[delta].sup.13]C Microphytobenthos -13.7 [+ or -] 1.6 (2) Marine POM -22.0 [+ or -] 1.3 (2) Riverine POM -28.2 [+ or -] 1.1 (2) Clams (control) -17.9 [+ or -] 0.2 (4) Clams (treatment) -17.8 [+ or -] 0.5 (4) [[delta].sup.15]N Microphytobenthos 8.9 [+ or -] 1.6 (2) Marine POM 4.6 [+ or -] 1.6 (2) Riverine POM 9.5 [+ or -] 0.6 (2) Clams (control) 11.3 [+ or -] 0.3 (4) Clams (treatment) 11.4 [+ or -] 0.5 (4) Sample October [[delta].sup.13]C Microphytobenthos -13.8 [+ or -] 0.5 (2) Marine POM -21.9 [+ or -] 0.8 (2) Riverine POM -30.4 [+ or -] 2.1 (2) Clams (control) -17.8 [+ or -] 0.9 (5) Clams (treatment) -17.8 [+ or -] 0.3 (4) [[delta].sup.15]N Microphytobenthos 9.7 [+ or -] 0.1 (2) Marine POM 5.6 [+ or -] 0.5 (2) Riverine POM 10.0 [+ or -] 0.9 (2) Clams (control) 11.3 [+ or -] 0.6 (4) Clams (treatment) 11.2 [+ or -] 0.3 (4) Sample December [[delta].sup.13]C Microphytobenthos -13.9 [+ or -] 0.1 (2) Marine POM -20.7 [+ or -] 0.9 (2) Riverine POM -31.3 [+ or -] 0.3 (2) Clams (control) -16.9 [+ or -] 0.8 (4) Clams (treatment) -17.9 [+ or -] 0.3 (4) [[delta].sup.15]N Microphytobenthos 8.9 [+ or -] 0.7 (2) Marine POM 5.6 [+ or -] 1.0 (2) Riverine POM 8.3 [+ or -] 0.2 (2) Clams (control) 11.0 [+ or -] 0.7 (4) Clams (treatment) 11.0 [+ or -] 0.2 (4) February Sample 2001 [[delta].sup.13]C Microphytobenthos -14.9 [+ or -] 0.8 (2) Marine POM -20.2 [+ or -] 0.9 (2) Riverine POM -26.4 [+ or -] 1.1 (2) Clams (control) -17.3 [+ or -] 0.3 (5) Clams (treatment) -17.0 [+ or -] 0.7 (5) [[delta].sup.15]N Microphytobenthos 9.7 [+ or -] 0.8 (2) Marine POM 7.0 [+ or -] 0.4 (2) Riverine POM 8.3 [+ or -] 1.0 (2) Clams (control) 11.0 [+ or -] 0.4 (4) Clams (treatment) 10.1 [+ or -] 0.6 (4) Sample April [[delta].sup.13]C Microphytobenthos -17.4 [+ or -] 0.6 (2) Marine POM -20.7 [+ or -] 1.5 (2) Riverine POM -24.4 [+ or -] 0.2 (2) Clams (control) -17.2 [+ or -] 0.3 (4) Clams (treatment) -17.5 [+ or -] 0.2 (4) [[delta].sup.15]N Microphytobenthos 7.6 [+ or -] 0.7 (2) Marine POM 6.7 [+ or -] 0.1 (2) Riverine POM 8.6 [+ or -] 0.3 (2) Clams (control) 10.7 [+ or -] 0.4 (3) Clams (treatment) 10.3 [+ or -] 0.4 (4) Sample June [[delta].sup.13]C Microphytobenthos -13.2 [+ or -] 1.1 (2) Marine POM -20.8 [+ or -] 0.2 (2) Riverine POM -26.0 [+ or -] 0.1 (2) Clams (control) -17.4 [+ or -] 0.7 (4) Clams (treatment) -18.0 [+ or -] 0.6 (4) [[delta].sup.15]N Microphytobenthos 8.4 [+ or -] 0.5 (2) Marine POM 3.7 [+ or -] 0.1 (2) Riverine POM 9.1 [+ or -] 0.2 (2) Clams (control) 9.2 [+ or -] 0.7 (4) Clams (treatment) 9.3 [+ or -] 0.3 (3) Sample Total [[delta].sup.13]C Microphytobenthos -14.5 [+ or -] 1.6 (12) Marine POM -21.1 [+ or -] 1.1 (12) Riverine POM -27.8 [+ or -] 2.7 (12) Clams (control) -17.4 [+ or -] 0.6 (26) Clams (treatment) -17.6 [+ or -] 0.6 (25) [[delta].sup.15]N Microphytobenthos 8.9 [+ or -] 0.9 (12) Marine POM 5.7 [+ or -] 1.3 (12) Riverine POM 9.0 [+ or -] 0.8 (12) Clams (control) 10.8 [+ or -] 0.8 (23) Clams (treatment) 10.5 [+ or -] 0.8 (23)