This study was undertaken to explore the applicability of glutathione S-transferase (GST) activity as a predictable indicator to monitor chemical pollution in shells and fishes utilized for food. There were some variations in the basal level of GST activity depending on species tested. Ark shells, Anadara satowi, showed the highest normal enzyme activity, followed by catfish and marine mussels, Mytilus coruscus. White clams, Meretrix lusoria, Israeli carp and catfish had lower activity. When A. satowi was exposed to 3-methyl-cholanthrene (3-MC), a prototypic polycyclic aromatic hydrocarbon for 1 week, GST activity decreased by about 30%. This reduction in GST activity induced by 3-MC did not recover until two weeks after the cessation of exposure. GST activity increased in response to 3-MC in most of the other species studied. The GST elevation in M. coruscus attained its maxinum of about 200% at the termination of 3-MC exposure maintaining this level up to 2 weeks, and declined gradually thereafter. 3-MC also induced GST activity in lsraeli carp in a similar fashion to M. coruscus. Phenobarbital induced GST activity both in M. coruscus and lsraeil carp. Other chemicals. such as clofibrate, butylated hydroxyanisole. hexachlorobenzene, and oxolinic acid did not change the enzyme activity significantly in most speciel. Phenol depressed GST activity only in lsraeli carp. These results suggest that the basal level of GST activity is somewhat variable and that the direction of change in response to chemicals seems to be related to its normal activity. The change in enzyme activity can be a preditable indicator of some environmental chemicals such as PAHs and phenol.