As observed in this survey, many contaminants co-vary with each other in the sediments. Therefore, 

 correlation analyses alone do not provide great insight into the potential causes of toxicity. A much 

 stronger weight of evidence is provided by the complementary measures of correlative strength, con- 

 centrations gradients between toxic and nontoxic samples, and comparisons with applicable effects- 

 based, numerical guidelines. 



CONCLUSIONS 



This survey was intended to provide information on the possible biological effects of toxic chemicals 

 in the sediments of the Hudson-Raritan estuary. Standardized laboratory toxicity tests were performed 

 on 174 samples collected throughout the study area. Some of the important conclusions derived from 

 this survey follow: 



Potential for Toxicity 



• The concentrations and mixtures of toxicants quantified in previous studies differed among the 

 many different waterways, tributaries, harbors, and basins of this study area. 



• The concentrations of many substances quantified during previous studies exceeded the concentra- 

 tions previously associated with toxicity, occasionally by a great amount, and therefore, suggested that 

 sediments in this area may be toxic. 



• The concentrations of polynuclear aromatic hydrocarbons (PAHs) were extremely high in some 

 samples from the East River collected during the present survey. The concentrations of chlorinated 

 hydrocarbons, such as PCBs, pesticides, and dioxins, were very high in some samples from the lower 

 Passaic River and Newark Bay. The concentrations of total simultaneously extracted metals exceeded 

 the acid-volatile sulfide concentrations in a few of the samples. 



• Based upon historical data, those areas included in the present survey in which the highest toxicity 

 was predicted included Newark Bay and Arthur Kill. As expected, many samples from these two 

 adjacent areas were highly toxic. Portions of the East River and lower Passaic River, which were 

 expected to be moderately toxic, often were moderately to highly toxic in the laboratory tests. 



Incidence of Toxicity 



The significance of the toxicity data was determined in statistical comparisons of the test results with 

 the respective controls. 



• Out of 58 sediment samples tested in previous studies, 45 (77.6%) were highly toxic to either 

 nematode growth or amphipod survival. 



• All four test end-points measured in the present survey indicated results significantly different 

 from controls in samples collected throughout the estuary. 



• Of the 117 stations that were sampled in Phase 1 of the present survey, test results for 81 stations 

 (69% of the total tested) were significantly different from controls in at least one of the test end-points. 



• Of the 117 samples, 46% were significantly different from controls in the amphipod survival tests. 



• Of 109 samples tested, 27% were significantly different from controls in the tests of bivalve em- 

 bryo survival or normal development. 



• Of 1 16 samples tested, 41% were significantly different from controls in the bacterial biolumines- 

 cence tests. 



• Of 57 samples from Newark Bay and vicinity tested during Phase 2, 48 (84%) were significantly 

 toxic to amphipod survival. 



126 



