In addition to implicating ammonia, Phase I testing also indicated 

 that toxicity in some samples was associated with non-polar organic 

 materials and with material that was retained on the filters. We 

 applied the following Phase II isolation techniques that were used in a 

 similar situation by Schubauer-Berigan and Ankley (1991). To verify 

 that toxicity was due in part to material retained on the filters, the 

 filters were extracted with methylene chloride. Filters used in Phase I 

 for samples from the Des Plaines River site DP277.0 and the Calumet Sag 

 Channel site CS307.4 were soaked in 10 ml of methylene chloride for 1 

 hour. The solvent was evaporated from the beakers and dilution water 

 was added to the same volume as the original filtered sample. The 

 extracts then were screened for acute toxicity. 



Having checked the toxicity of the material on the filters, we 

 next investigated the nonpolar organics using solid phase Cjg absorption 

 columns and subsequent chromatography. To maximize the extraction of 

 possible toxicants, filtration was omitted and porewater was centrifuged 

 at 10,000 g for 30 minutes to remove particles that would clog the Cjg 

 column. The supernatant from the centrifugation step was checked for 

 toxicity. If toxicity was present, a 200-ml sample of the supernatant 

 was passed over a 6-ml Cjg column that had been conditioned with 25 ml 

 of methanol followed by 25 ml of Millipore^^ ultrapure water. Post 

 column aliquots were collected after passage of 25 ml and 100 ml of 

 methanol and tested for toxicity. 



Toxicity was not recovered from the DP277.0 sample using 100% 

 methanol elutions of the C^g columns as suggested by Mount and Anderson- 

 Carnahan (1989), so we eluted the columns with increasingly nonpolar 

 mixtures of methylene chloride in methanol (1, 5, 10, 15, 20, 25, 30, 

 35, 40, 45, 50, and 100%) as suggested by Schubaurer-Berigan and Ankley 



22 



