The mean mortality in all locations exceeded that of the Yaquina Bay controls. Sediments from eastern 

 Pralls Island, the Hackensack River, northwest Pralls Island and Sawmill Creek caused the highest 

 mortality. The results within each of the areas were highly variable, except in the sand sediments from 

 West Bayonne. Lowest mortality was observed in the sandy sediments from Isle of Meadows and 

 Constable Hook. In all three cases in which sand and mud from the same general location were tested, 

 mean mortality was higher in the mud. 



In summarizing the data from these small, disparate surveys, several patterns in toxicity seem to emerge. 

 The data from the two Aqua Survey, Inc. surveys generally agree with those from Tietjen and Lee 

 (1984), Scott et al. (1990), the City of New York (Brosnan and O'Shea, 1993), and EPA's EMAP 

 surveys (Schimmel et al., 1994). That is, all five studies indicated that sediments from the Arthur Kill, 

 Newark Bay, the lower Passaic River, and Kill van Kull were highly toxic. Moreover, all five studies 

 indicated that samples from the northern reach of Arthur Kill (from Isle of Meadows to Shooters Is- 

 land) were particularly toxic. Several of the studies indicated that samples from the lower Hudson 

 River and the upper New York Harbor were not toxic. Also, toxicity to amphipods generally dimin- 

 ished eastward into the western Long Island Sound. Some samples from the lower reaches of the 

 estuary and inner New York Bight were toxic in at least one test. 



Numerous small studies of sediment toxicity have been conducted by or for the Army Corps of Engi- 

 neers throughout the estuary as requirements for ocean disposal dredging permits. These toxicity tests 

 were performed with consistent protocols, and, together, provide internally comparable data for much 

 of the study area. Usually, one to several samples were tested in each study, each consisting of sedi- 

 ments collected in one or more long sediment cores. Data were available and compiled from 76 reports 

 (Public Notices for Dredged Material Disposal, U. S. Army Corps of Engineers, 1985-1993). Tests 

 were conducted with the shrimp Palaemonetes pugio, the polychaete Nereis virens, and the clam 

 Mercenaria mercenaria exposed to solid phase sediments for 10 days, using U.S. EPA/ACOE (1977) 

 protocols. Test results were compared to those from an offshore reference site to determine significant 

 toxicity. Sediments were tested from shipyards, marine terminals, industrial waterways, harbors, sew- 

 age outfalls, navigation channels, barge berthing sites, petroleum docks, military docks, tributary riv- 

 ers, and creeks in all of the major regions of the estuary (Figure 3). 



Ninety-two samples were tested with the three bioassays, for a total of 276 individual tests. The major- 

 ity of the samples did not cause significantly elevated mortality in the test organisms. However, mor- 

 tality significantly different from the reference materials was indicated in sediments from the lower 

 Raritan River, Arthur Kill, Kill van Kull, Keyport Harbor in Raritan Bay, Gowanus Bay/Canal, and the 

 upper East River near Riker's Island (Figure 3). The samples that were toxic in these tests were col- 

 lected in many of the same areas in which sediments either were estimated to be highly contaminated 

 (e.g., Squibb et al., 1991) or were toxic in tests performed with other taxa (Figure 2). 



Tatem et al. (1991) tested sediments from three sites — Westchester Creek adjoining the upper East 

 River, central Arthur Kill, and Gowanus Creek — for toxicity to the mysid, Mysidopsis bahia. The 

 sediments were held for differing periods of time, from 8 days to 40 weeks, before testing was initiated. 

 In the tests performed with sediments held for fewer than 8 days, the samples from Westchester Creek 

 and Arthur Kill were significantly more toxic than offshore reference samples. The Gowanus Creek 

 samples were not toxic. 



13 



