A comparison of mussels (Perna viridis) and semi-permeable membrane devices (SPMDs) was carried out at five sites, representing a gradient of contaminant concentrations, in Hong Kong coastal waters. Mussels, originally collected from a "clean" location, were deployed along with SPMDs at each site for 30 days. Analyses for chlorinated pesticides and polychlorinated biphenyls (PCBs) indicated that SPMDs have potential as monitoring tools, and to some extent can overcome the problems associated with mussels, such as natural variability, differing age, sex, and physical condition. However, in most cases, SPMDs failed to rank the sites in the same order as mussels in terms of contaminant concentrations. Nonetheless, in localities where mussels cannot survive--as shown at Kwun Tong in the present experiment--SPMDs may be valuable in providing an indication of potentially bio-available lipophilic pollutants.