additional procedures used parametric methods, the results cannot be used in any inferential 

 sense, but, rather, provide additional qualitative descriptions of the results. 



RESULTS 



Sediment Toxicity Test Results 



Solid Phase Bioassay with the Amphipod Rhepoxynius abronius. Results for three end- 

 points in this bioassay are summarized in Table 5. In the control sediments from the 

 amphipod collection site in Puget Sound, means of 19.6 ± 0.5 and 18.8 ± 1.3 survivors out of 

 20 (98 and 94 percent, respectively) were observed. Mean survival that was lower than the 

 control means was seen in tests of all the samples. Among the most toxic, as judged by the 

 fewest survivors, were samples 1, 2, 3, 7, 11, 13, and 15 in which 50 percent of the amphipods 

 or fewer survived. Based upon previous experience with this bioassay, these results are 

 indicative of highly toxic sediments. Usually, a result showing less than 75 percent 

 survival is significant relative to controls (Swartz et ah, 1985; Mearns et ah, 1986), and a 

 result showing fewer than 50 percent survivors is observed only in the most contaminated 

 sites (Swartz et ah 1985). Mean survival in sample 1 sediments was roughly one-third that 

 in sample 10 sediments. 



Means of 1.0 ± 2.2 and 1.4 ± 1.7 amphipods avoided (emerged from) the control sediments. 

 Mean avoidance of the test sediments was relatively similar to that in controls in only three 

 samples: numbers 3, 6, and 13. Mean avoidance of sediment samples 7, 8, and 9 was roughly 5 

 times higher than avoidance of controls. However, within-sample variability was 

 relatively high for this end-point. Percent reburial of survivors at the end of the tests was 

 similar in all samples and this end-point was not evaluated further. 



Solid Phase Toxicity Test with the Amphipod Ampclisca abdita. The toxicity data are 

 presented in Table 6 as both counts and percents of survivors. Occasionally, more than 20 

 animals were inadvertently initially exposed to the sediments, despite the sieving step 

 performed before the tests to remove native A. abdita. Means of 17.2 and 16.0 individuals (86 

 and 80 percent, respectively) survived the 10-d flow-through test in control sediments. The 

 lowest mean percent survival occurred in samples 1, 2, and 3. Mean percent survival was 

 highest in the SP and TB sediments. In some samples, i.e., those from SP, the amphipods 

 showed higher mean survival than those exposed to the sediment from the amphipod 

 collection site. Mean survival in sample 14 exceeded that for sample 1 by a factor of 1.4. 

 Mean cumulative avoidance of the test sediments under flow-through conditions was highest 

 in the OA sediments followed, in order, by avoidance of Control 1, VA, YB, TB, and SP 

 sediments. Mean avoidance of OA sediments exceeded that of TB by a factor of about 3. 



These tests of all 15 samples were performed in flow-through conditions, whereas those 

 with R. abronius were performed under static conditions. However, three samples also were 

 tested with A. abdita under static conditions identical to those for R. abronius to provide a 

 comparable basis for evaluation of the results. Mean survival was slightly lower in sample 1 

 and controls under static test conditions than in flow-through conditions, but was unchanged in 

 samples 4 and 13. 



Elutriate Toxicity Test with Embryos of the Mussel Mytilus edulis. Data for the two end- 

 points of this test are summarized in Table 7. Both end-points indicated that the three OA 

 samples were the most toxic of all 15 stations tested. About 94 and 95 percent normal larvae 

 were observed in sediment controls and seawater controls, respectively, compared to a site 

 mean of 75 percent for OA. Relative to the controls, sediments from all stations caused lower 

 percentages of normal development. Also, relative to controls, mean percent survival was 

 lower in embryos exposed to samples from all stations. About 76 and 100 percent of the 

 larvae survived in sediment controls and seawater controls, respectively. As few as 29.2 

 percent survived in OA samples. Mean percent normal development was roughly 1.3 times 

 lower for sample 3 than for sample 8 and for the sediment control. Mean percent survival 

 was about 2.5 times higher in sample 8 samples and the sediment control than in sample 3. 



30 



