Abstract :
Metal concentrations in individual oysters were estimated in order to assess the among-individuals metal variability in a geographically confined area. American oysters (Crassostrea virginica) were collected from three sites on the Potomac River, Maryland, and the soft tissues of individuals was analysed for Ag, Cd, Cu, Fe, Ni, Pb and Zn. At all sites metal concentrations were highly variable with the coefficient of variation (CV) between 15 and 55%, the three site mean increasing as follows: Fe; Cd; Ni; Zn; Pb; Ag; Cu. Excluding Pb, the analytical contribution to the total variance was about 2%. Bootstrap analysis, a statistical resampling procedure with replacement, was used to estimate population parameters for pooled samples. Simulated sampling of 10 000 pooled samples verified that the sampling error decreases significantly with increasing numbers of individuals per pool. The confidence interval of means for samples consisting of 20 oysters varied from ±5% to ±20% (p < 0.05), depending on metal and site. The confidence interval of means increased by a factor of approximately 1.5 when 10 individuals were used instead of 20, and decreased by a factor of 0.6 when 50 oysters were used per pooled sample. These results indicate that if analyses of individuals are precluded, increased size of pooled samples is a low cost option for increasing the ability of monitoring programmes to detect changes in environmental contaminants.
