The most common result in Table 5 is a lack of trends. Among the 2156 combinations of 14 chemicals 

 at 154 sites, there are only 41 increases and 217 decreases at the 95% level of confidence. On a per- 

 chemical basis the trends, with 95% confidence, are as follows: 



Given a 5% probability of random data showing trends, there could be 54 increases and 54 decreases 

 that are not real trends. Conceivably, none of the 41 increases are real. The important point, however, is 

 that decreases gready outnumber increases. Decreases exceed increases by a factor of three or more for 

 all the chlorinated hydrocarbons, tributyltin, arsenic, cadmium, copper, and selenium. 



Annual data have been aggregated to national geometric means in Table 6 (O'Connor, 1995) where, 

 because of species differences, concentrations in mussels have been separated from those in oysters for 

 Cu, Pb, and Zn. Decreasing trends at the 95% confidence level and at this national level of aggregation 

 exist for As, Cd, Cu (in mussels), all the chlorinated organics, and SBT. Except for selenium, this list 

 highlights the same decreasing trends found by counting trends site-by-site. National geometric means 

 show trends for copper in mussels but not oysters, which is consistent with the site-by-site data where 

 16 of the 17 decreases were at mussel-sites. 



Decreasing trends are not unexpected. All the monitored chlorinated hydrocarbons have been banned 

 for use in the United States and tributyltin has been banned as a biocide on recreational boats. For 

 cadmium and arsenic there have been decreases in their uses. 



Annual consumption of Cd in the U.S. averaged 4069 metric tons per year (mt y-i) for 1986 through 

 1989, dropped by 25% in 1990, and averaged 3068 mtyi for 1990 through 1993 (DOI, 1994). While 

 the use of cadmium in rechargeable batteries is growing, the dramatic decrease in overall use is due to 

 drops in uses for electroplating onto metal surfaces for rustproofmg automobile parts, as a stabilizer in 

 plastics, and its use in pigments. Annual consumption of arsenic in the U.S. has remained fairly steady 

 at around 22,000 mt yi since 1986. Its primary use is as an ingredient in wood preservatives, which has 

 remained fairly stable, but there has been a decline in its agricultural uses. Between 1986 and 1991, the 

 annual agricultural use varied between 4200 and 5000 mt y ^ but in 1992 and 1993, respectively, only 

 3900 and 3000 mt y-i of arsenic were used. There has been no parallel decrease in use of copper within 

 the United States. Possibly, its decreasing trend indicates increased success in control of copper emis- 

 sions from many sources. 



While concentration data themselves do not indicate whether trace element levels in moUusks are 

 affected by human activity, the existence of monotonic trends like those identified here are likely due to 

 activities that are increasing or decreasing an element's concentration in the environment. All sorts of 

 natural factors, either internal, such as moUusk growth or reproduction, or external, such as tempera- 



