EXECUTIVE SUMMARY 



An evaluation of the response and sensitivity of candidate measures of biological effects 

 to a range in contaminant concentrations was performed in the San Francisco Bay area in 

 1987. The evaluation was performed to determine which, if any, of the candidate measures 

 may be useful in the National Status and Trends (NS&T) Program of the National Oceanic 

 and Atmospheric Administration (NOAA). 



The NS&T Program analyzes three media— sediments, bottomfish, and bivalves— 

 routinely at sites nationwide. The present evaluation included biological tests of two of 

 those media: sediments and fish. The overall approach chosen for the evaluation involved 

 analyses of samples collected in the field at sites that were presumed to represent a range in 

 chemical contamination. The biological tests were performed with subsamples of samples 

 that were also analyzed for chemical concentrations. All the tests were performed "blind," 

 i.e., without knowledge of the origin of the samples. Data from the various biological tests 

 were then compared with each other and with the chemical data in various statistical 

 procedures. It was presumed and hypothesized before the evaluation begarT^hat biological 

 tests most applicable to the NS&T Program would be those that were able to indicate 

 differences among sampling locations over a range in chemical contamination and/or between 

 sampling locations and laboratory controls, had relatively large ranges in response among 

 mean values for the sampling locations, had relatively small analytical errors, and 

 indicated patterns in biological response that generally paralleled the pattern in chemical 

 contamination. 



The relative sensitivity, analytical precision, discriminatory power, and concordance 

 among end-points and with sediment chemistry were compared among multiple end-points of 

 five types of sediment toxicity tests. The tests were performed with aliquots of 15 

 composited, homogenized samples collected in San Francisco Bay and Tomales Bay, 

 California. Each sample was also tested for trace metal and organic compound concentration, 

 organic carbon content, and texture. The end-points evaluated were: survival and avoidance 

 of solid phase sediments by the amphipods Rhepoxynius abronius and Ampelisca abdita; 

 survival and abnormal development in the embryos of the mussel Mytilus edulis exposed to 

 elutriates; fertilization success, abnormal development, echinochrome pigment content, 

 incidences of abnormal mitotic division, micronuclei, cytological abnormalities and mitoses 

 per embryo in the embryos of the urchin Strongylocentrotus purpuratus exposed to elutriates; 

 and survival and egg production in the polychaete Dinophilus gyrociliatus exposed to 

 interstitial (pore) water. Among the end-points evaluated, abnormal development of M. 

 edulis embryos was the most sensitive to the samples relative to controls and had the 

 highest precision and discriminatory power. Survival of R. abronius was the second most 

 sensitive and also had a high range in response and discriminatory power. The results of 

 both end-points (along with those of M. edulis survival), however, were more highly 

 correlated with sedimentological variables than with the concentrations of chemical 

 contaminants. The end-point of A. abdita survival had relatively high analytical precision, 

 moderate discriminatory power and was relatively highly correlated with several 

 chemicals, but had comparatively low sensitivity relative to controls. Abnormal 

 development and echinochrome content in S. purpuratus had relatively high precision and 

 results were relatively highly correlated with several chemicals, but discriminatory power 

 was moderate and the abnormal development results contradicted those of many of the other 

 end-points. Several of the cytological/cytogenetic end-points of this test measured in only 

 five samples indicated a wide range in response and strong correlations with chemical data, 

 but precision was relatively low. The test of D. gyrociliatus egg production was intermediate 

 in sensitivity, had relatively low precision and discriminatory power, and was highly 

 correlated with several organic chemical groups. The results of this pore water test were not 

 highly correlated with those of the solid phase and elutriate tests. The authors conclude 

 that, since different toxicological mechanisms may occur in the responses of organisms to 

 complex media such as sediments, multiple toxicity tests are needed to comprehensively 

 assess the quality of marine sediments. 



