allows us to extrapolate from chemical 

 data describing the nationwide distribu- 

 tion of contamination to give us a sense 

 of the distribution of sediment toxicity. 

 They reviewed 1 50 scientific papers and 

 reports on sediment toxicity and found 

 85 with data on both biological response 

 and chemical concentrations in the 

 sediment. They examined all of this in- 

 formation and found chemical con- 

 centrations, listed in Table 3, that corre- 

 spond to concentrations above which 

 effects were frequently observed. Since 

 the values in Table 3 (except for chro- 

 mium) are all greater than those defined 

 in this report as "high" concentrations, it 

 follows that sediment toxicity will be 

 found at fewer sites than are highlighted 

 in the Appendix and Figure 2. The di- 

 chotomy between "high" and "possibly 

 biologically effective" concentrations is 

 actually somewhat larger because the 

 sediment data used by Long and Mor- 

 gan were not adjusted for their sand 

 content. The middle column in Table 3 

 shows that the "high" NS&T concentra- 

 tions would have been about 40% lower 

 if the data had not been adjusted. 



As noted in Table 3, the exception of 

 chromium is probably due to its naturally 

 high levels in the Northwest combined 

 with the fact that many tested sediments 

 have been collected in that region. As 

 Long and Morgan indicated, chromium 

 probably was not causing toxicity in any 

 of the tested sediments. 



This extrapolation from chemistry to 

 biological effect cannot be accepted 

 unequivocally. The exhaustive compi- 

 lation of reports that Long and Morgan 

 used included bioassays based on dif- 

 ferent organisms, different test proce- 

 dures, and different indicators of effect 

 (usually death). The list of reports in- 



cluded studies where chemicals were 

 added to sediments, cases where sedi- 

 ments were tested as taken from the 

 field, and cases where toxicity was esti- 

 mated from calculations of chemical 

 concentrations in the pore water of 

 sediments. Finally, it included reports 

 where sediment quality was judged on 

 the basis of the species of organisms 

 found (or not found) living in association 

 with it. All of these differences make 

 comparisons precarious. Nevertheless, 

 the conclusion that sediment toxicity is 

 not widespread is consistent with re- 

 sults from NS&T studies that tested 

 sediment toxicity. 



Scott (1 989) collected sediment at NS&T 

 sites in western Long Island Sound and 

 in the Hudson/Raritan estuary that are 

 listed in the Appendix as having sedi- 

 ment with high levels of contamination. 

 In a standard bioassay based on 1 0-day 

 survival of amphipods these sediments 

 proved nontoxic. It may be that toxic 

 sediments are found in only very local- 

 ized and highly contaminated places. 

 For example, Rodgerson et al. (1985), 

 using that same amphipod test, did find 

 toxicity in Black Rock HarlDor near Bridge- 

 port, CT, a site already mentioned as 

 being highly contaminated. 



While the NS&T sediment data do not 

 indicate high levels of contamination in 

 San Francisco Bay, Chapman et al. 

 (1987) and Long and Buchman (1989) 

 did find toxic responses to sediments 

 taken from Islais Waterway and the inner 

 parts of Oakland Harbor, two highly 

 industrialized locations within San Fran- 

 cisco Bay. Swartz et al. (1982) found a 

 range of toxic responses within the in- 

 dustrial waterways of Tacoma, WA, off 

 Commencement Bay. There is no 

 question that sediment toxicity exists in 



10 



