the nontoxic samples (often by a considerable amount), and they exceeded the applicable guidelines 

 (Table 40). The concentrations of the low molecular weight PAHs, in particular, corresponded very 

 well with the toxicity results. The concentrations of the petroleum-related compounds were very high 

 in the samples that were toxic in the Microtox tests, whereas the combustion-related compounds were 

 relatively high in the samples that were toxic in the amphipod tests. The concentrations of acenaphthene 

 were correlated with toxicity, and were highly elevated in the toxic samples, but they exceeded the 

 applicable SQC by only a factor of 2.8 and 4.8, respectively, in the two tests. Also, phenanthrene was 

 correlated with toxicity and elevated in concentration in the toxic samples; in addition, it was elevated 

 relative to the applicable SQC by factors of 10.4 and 17.0, respectively, in the two tests. 



In summary, these data from the Phase 1 portion of the survey suggest that the toxicity observed in the 

 samples was in strong correspondence with the concentrations of the PAHs in the samples. To a con- 

 siderably lesser degree, toxicity corresponded with the concentrations of some trace metals and chlori- 

 nated organic compounds. 



In the samples analyzed during Phase 2 of the survey, a considerable number of chemicals co-varied 

 with amphipod survival (Table 41). Nine trace elements were correlated with toxicity to the amphi- 

 pods; however, the concentrations of most of these metals were not highly elevated relative to appli- 

 cable guidelines. For example, cadmium, chromium, and copper were correlated with amphipod sur- 

 vival, but the concentrations of these substances were below the ERM values. The concentrations of 

 mercury were elevated relative to the ERM values, but Long et al. (1995) reported only a moderate 

 degree of confidence in the guidelines for mercury. 



Whereas the concentrations of PAHs were highly correlated with amphipod survival in Phase 1, the 

 concentrations of PCBs, 2,3,7,8-tcdd, and many other chlorinated hydrocarbons were highly correlated 

 with amphipod survival in Phase 2 (Table 41). The strong correlations with are particularly interesting 

 since dioxins generally are not especially toxic to invertebrates in short-term acute exposures. No 

 applicable guidelines are available for many of these compounds. The concentrations of dieldrin were 

 far below the applicable sediment guideline values, whereas the concentrations of the p,p'-DDE and 

 total DDT isomers were high relative to the ERM values of Long et al (1995), but were far below the 

 respective SECs of Mac Donald (1994). 



The PAH concentrations in some samples exceeded the sediment guidelines of Long et al. (1995) or the 

 U.S. EPA (1994) and the average concentrations in the toxic samples exceeded the average in the 

 nontoxic samples. However, amphipod survival was not correlated with those compounds, and, there- 

 fore, they were not included in Table 41. Nevertheless, they may have been important in contributing 

 to toxicity in some specific samples in which the PAH concentrations were particularly high. 



In previous studies conducted within this study area, measures of sediment toxicity were highly corre- 

 lated with a number of different chemicals. Scott et al. (1990) reported that amphipod mortality in 

 samples from the Hudson-Raritan Estuary was correlated with the concentrations of total PCBs, total 

 PAHs, several pesticides, copper, zinc, chromium, lead, nickel, and cadmium. Also, the concentrations 

 of many of these chemicals in the highly toxic samples equalled or exceeded the respective ERM 

 values. The correlations between amphipod mortality and the concentrations of trace metals normal- 

 ized to the aluminum content were highly significant in the EMAP samples (Schimmel et al., 1994). In 

 samples collected by the City of New York (Brosnan and O'Shea, 1994), amphipod mortality was 

 correlated with total SEM/AVS ratios. 



124 



