Test procedures. Before toxicity tests were conducted, the pore water samples were 

 measured and adjusted, if necessary, to produce water with temperature of 20° ± 1°C, 

 ammonia concentration of <2mg/L, salinity of 25 ± 1 ppt, pH of 8.0 ± 0.2, and dissolved 

 oxygen of > 80 percent saturation (Battelle, 1987a). Following water quality adjustments, the 

 life-cycle toxicity test with Dinophilus gyrociliatus was conducted to determine mortality and 

 sublethal reproductive effects (i.e., eggs laid per female), using the procedures of Carr et al. 

 (1986; in press; Battelle, 1987b). The tests were conducted in 20-mL Stender dishes with 

 ground glass lids, with 10 mL of pore water per dish. At a minimum, four animals were 

 placed into each dish, following the addition of the pore water. The tests were started with 

 1- to 2-d old animals. Because the worms reach mature size rapidly, an experienced 

 investigator can easily identify newly released juveniles by their small size. Survival and 

 reproductive data for each chamber were recorded after 1, 4, and 7 days. The reproductive 

 data recorded for each chamber consisted of the total number of female eggs, the number of 

 egg cases, and the number of newly emerged juveniles. All observations and manipulations 

 were performed using a dissecting microscope with fiber optic illumination. The test animals 

 were fed 50 ul of a 0.5 percent spinach food suspension in each dish. Tests were performed in 

 two batches: the first consisting of samples from the TB and SP sites; and the second 

 consisting of the samples from the VA, YB, and OA sites. A control pore water sample from 

 Duxbury Bay, Massachusetts was tested concurrently with each batch of samples. 



Sediment Chemical Analyses 



Chemical and texture analyses of the sediments were performed by SAIC. 



Organic Compounds. The methods were based upon the protocols of the NOAA NS&T 

 Program (MacLeod et al., 1985). A 50-g (ww) sediment sample was divided into aliquots for 

 analyses of organic compounds. Internal standards (to permit assessment of analyte recovery 

 efficiency) were spiked into the sample, which was extracted four times with varying 

 amounts of methanol then dichloromethane. A bottle roller or shaker table was used to mix 

 the sample with the extraction solvents. The combined extracts were concentrated to about 1 

 mL in a Kuderna-Danish apparatus and solvent-exchanged into hexane in preparation for 

 fractionation. 



The concentrated extract was loaded onto a precalibrated chromatography column packed 

 with silica gel, alumina, and granular copper. A first fraction (SA1) was eluted with 

 hexane, the second fraction (SA2) with 50 percent dichloromethane in hexane and the third 

 fraction (SA3) with dichloromethane and methanol. Fraction SA1 (containing saturated 

 hydrocarbons) was concentrated for analysis by gas chromatographic-electron capture detector 

 (GC-ECD). Fraction SA2 (containing aromatic and chlorinated hydrocarbons) initially was 

 concentrated and then loaded onto a precalibrated column packed with Sephadex LH-20. A 

 subtraction (SA2-L1) was eluted with a 6:4:3 mixture of cyclohexane-methanol- 

 dichloromethane. This fraction, containing biogenic material, then was concentrated for 

 portion analysis by GC-ECD. A second subfraction (SA2-L2) was eluted with the same solvent 

 mixture. Fraction SA2-L2 was concentrated again, solvent-exchanged into hexane, further 

 concentrated under a stream of nitrogen, and spiked with an additional internal standard (to 

 allow correction for instrument injection volume fluctuations) in preparation for instrument 

 analysis. Fraction SA3 (containing coprostanol, a natural enteric product selected for analysis 

 as a sewage tracer) was concentrated, solvent-exchanged, concentrated again, and spiked with 

 an additional internal standard as described for the SA2-L2 fraction 



Fraction SA2-L2 was analyzed for polynuclear aromatic hydrocarbons (PAHs) by gas 

 chromatographic-flame ionization detector (GC-FID) and for chlorinated hydrocarbons by GC- 

 ECD. If hexachlorobenzene (HCB) was found in SA2-L2, fraction SA1 also was analyzed (for 

 additional HCB) by GC-ECD. 



The method detection limits attained in the analyses of organics in sediments are listed 

 in Tables 1 and 2. 



12 



