Another source of variability in results among the tests is the exposure medium that was tested. The 

 amphipods are tube-dwelling, epifaunal organisms and were exposed to solid phase sediments for 10 

 days. The clam larvae were planktonic and exposed to liquid phase samples for 48 hours. The bacteria 

 were cultured in laboratory equipment and exposed for 5 minutes to an organic extract of the sedi- 

 ments. The solvent extraction elutes potentially toxic contaminants from the sediments, and, therefore, 

 enhances their apparent availability. The Microtox tests probably provide an estimate of the toxicity 

 potential of the total bulk contaminant content of the sediments, instead of the biologically available 

 fraction. 



Summary of Chemistry/Toxicity Relationships. The data from Phase 1 of the Hudson-Raritan Estu- 

 ary survey were examined to determine (1) which chemicals were correlated significantly with the 

 measures of toxicity; (2) which chemicals also were elevated in concentration in the highly toxic samples 

 versus the nontoxic samples; and (3) which chemicals in the highly toxic samples also were elevated in 

 concentration above applicable, effects-based, sediment quality guidelines. A summary of the data 

 from these three tiers of analyses are listed in Table 40. The substances included in Table 40 are those 

 that were significantly correlated with two measures of toxicity: amphipod survival and microbial 

 bioluminescence. There were no strong, significant correlations between toxicity to the bivalve em- 

 bryos and any of the chemicals or chemical classes, so the data from those tests were not included in 

 Table 40. 



Amphipod survival was significantly correlated with two metals, mercury and tin, in the sediments. 

 Microtox test results were significantly correlated with 10 metals. Generally, the average concentra- 

 tions of all 10 trace metals in the highly toxic samples were slightly higher than in the nontoxic samples 

 in both of the toxicity tests. For most of the metals, the average concentrations in the highly toxic 

 samples were lower than the respective ERM values of Long et al. (1995). The concentrations of 

 mercury, lead, and zinc in the highly toxic samples either equalled or slightly exceeded the applicable 

 ERM values. Thirty of the 38 samples equalled or exceeded the mercury ERM concentration. How- 

 ever, Long et al. (1995) reported that they had only a moderate degree of confidence in the ERM value 

 for mercury, whereas they reported a relatively high degree of confidence in the values for lead and 

 zinc. Therefore, the exceedances of the mercury ERM concentration in most samples probably were 

 not very meaningful. 



Four chlorinated organic compounds were significantly correlated with microbial bioluminescence 

 and one was correlated with amphipod survival (Table 40). No applicable guidelines were available 

 for cis-chlordane and trans-nonachlor. Both of these isomers of chlordane were slightly elevated in the 

 highly toxic samples relative to the nontoxic samples. The concentrations of p,p'-DDT were signifi- 

 cantly correlated with both measures of toxicity. However, the average concentrations of this isomer of 

 DDT were much more elevated in the samples that were toxic to the amphipods (exceeded the ERM by 

 a factor of 21.6) than in those that were toxic to microbial bioluminescence (exceeed the ERM by a 

 factor of 1 .7). Long et al. (1995) reported a moderate degree of confidence in the ERM values for total 

 DDT and p,p'-DDE and Long and Morgan (1990) reported a low degree of confidence in the ERM for 

 p,p'-DDT Therefore, the slight exceedance of the ERM value for p,p'-DDT in the samples that were 

 highly toxic to microbial bioluminescence probably is meaningless. The concentrations of the DDT 

 isomers in these samples were one to two orders of magnitude lower than the respective Sediment 

 Effect Concentrations (SECs) of MacDonald (1994). 



All of the compounds and classes of PAHs were significantly correlated with the results of the amphi- 

 pod and Microtox tests the concentrations in the highly toxic samples exceeded the concentrations in 



122 



